online learning covid 19 essay

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Review Article
• Published: 27 September 2021

Why lockdown and distance learning during the COVID-19 pandemic are likely to increase the social class achievement gap

• Sébastien Goudeau   ORCID: orcid.org/0000-0001-7293-0977 1 ,
• Camille Sanrey   ORCID: orcid.org/0000-0003-3158-1306 1 ,
• Arnaud Stanczak   ORCID: orcid.org/0000-0002-2596-1516 2 ,
• Antony Manstead   ORCID: orcid.org/0000-0001-7540-2096 3 &
• Céline Darnon   ORCID: orcid.org/0000-0003-2613-689X 2  

Nature Human Behaviour volume  5 ,  pages 1273–1281 ( 2021 ) Cite this article

114k Accesses

148 Citations

129 Altmetric

Metrics details

The COVID-19 pandemic has forced teachers and parents to quickly adapt to a new educational context: distance learning. Teachers developed online academic material while parents taught the exercises and lessons provided by teachers to their children at home. Considering that the use of digital tools in education has dramatically increased during this crisis, and it is set to continue, there is a pressing need to understand the impact of distance learning. Taking a multidisciplinary view, we argue that by making the learning process rely more than ever on families, rather than on teachers, and by getting students to work predominantly via digital resources, school closures exacerbate social class academic disparities. To address this burning issue, we propose an agenda for future research and outline recommendations to help parents, teachers and policymakers to limit the impact of the lockdown on social-class-based academic inequality.

Similar content being viewed by others

Large socio-economic, geographic and demographic disparities exist in exposure to school closures

Elementary school teachers’ perspectives about learning during the COVID-19 pandemic

Uncovering Covid-19, distance learning, and educational inequality in rural areas of Pakistan and China: a situational analysis method

The widespread effects of the COVID-19 pandemic that emerged in 2019–2020 have drastically increased health, social and economic inequalities 1 , 2 . For more than 900 million learners around the world, the pandemic led to the closure of schools and universities 3 . This exceptional situation forced teachers, parents and students to quickly adapt to a new educational context: distance learning. Teachers had to develop online academic materials that could be used at home to ensure educational continuity while ensuring the necessary physical distancing. Primary and secondary school students suddenly had to work with various kinds of support, which were usually provided online by their teachers. For college students, lockdown often entailed returning to their hometowns while staying connected with their teachers and classmates via video conferences, email and other digital tools. Despite the best efforts of educational institutions, parents and teachers to keep all children and students engaged in learning activities, ensuring educational continuity during school closure—something that is difficult for everyone—may pose unique material and psychological challenges for working-class families and students.

Not only did the pandemic lead to the closure of schools in many countries, often for several weeks, it also accelerated the digitalization of education and amplified the role of parental involvement in supporting the schoolwork of their children. Thus, beyond the specific circumstances of the COVID-19 lockdown, we believe that studying the effects of the pandemic on academic inequalities provides a way to more broadly examine the consequences of school closure and related effects (for example, digitalization of education) on social class inequalities. Indeed, bearing in mind that (1) the risk of further pandemics is higher than ever (that is, we are in a ‘pandemic era’ 4 , 5 ) and (2) beyond pandemics, the use of digital tools in education (and therefore the influence of parental involvement) has dramatically increased during this crisis, and is set to continue, there is a pressing need for an integrative and comprehensive model that examines the consequences of distance learning. Here, we propose such an integrative model that helps us to understand the extent to which the school closures associated with the pandemic amplify economic, digital and cultural divides that in turn affect the psychological functioning of parents, students and teachers in a way that amplifies academic inequalities. Bringing together research in social sciences, ranging from economics and sociology to social, cultural, cognitive and educational psychology, we argue that by getting students to work predominantly via digital resources rather than direct interactions with their teachers, and by making the learning process rely more than ever on families rather than teachers, school closures exacerbate social class academic disparities.

First, we review research showing that social class is associated with unequal access to digital tools, unequal familiarity with digital skills and unequal uses of such tools for learning purposes 6 , 7 . We then review research documenting how unequal familiarity with school culture, knowledge and skills can also contribute to the accentuation of academic inequalities 8 , 9 . Next, we present the results of surveys conducted during the 2020 lockdown showing that the quality and quantity of pedagogical support received from schools varied according to the social class of families (for examples, see refs. 10 , 11 , 12 ). We then argue that these digital, cultural and structural divides represent barriers to the ability of parents to provide appropriate support for children during distance learning (Fig. 1 ). These divides also alter the levels of self-efficacy of parents and children, thereby affecting their engagement in learning activities 13 , 14 . In the final section, we review preliminary evidence for the hypothesis that distance learning widens the social class achievement gap and we propose an agenda for future research. In addition, we outline recommendations that should help parents, teachers and policymakers to use social science research to limit the impact of school closure and distance learning on the social class achievement gap.

Economic, structural, digital and cultural divides influence the psychological functioning of parents and students in a way that amplify inequalities.

The digital divide

Unequal access to digital resources.

Although the use of digital technologies is almost ubiquitous in developed nations, there is a digital divide such that some people are more likely than others to be numerically excluded 15 (Fig. 1 ). Social class is a strong predictor of digital disparities, including the quality of hardware, software and Internet access 16 , 17 , 18 . For example, in 2019, in France, around 1 in 5 working-class families did not have personal access to the Internet compared with less than 1 in 20 of the most privileged families 19 . Similarly, in 2020, in the United Kingdom, 20% of children who were eligible for free school meals did not have access to a computer at home compared with 7% of other children 20 . In 2021, in the United States, 41% of working-class families do not own a laptop or desktop computer and 43% do not have broadband compared with 8% and 7%, respectively, of upper/middle-class Americans 21 . A similar digital gap is also evident between lower-income and higher-income countries 22 .

Second, simply having access to a computer and an Internet connection does not ensure effective distance learning. For example, many of the educational resources sent by teachers need to be printed, thereby requiring access to printers. Moreover, distance learning is more difficult in households with only one shared computer compared with those where each family member has their own 23 . Furthermore, upper/middle-class families are more likely to be able to guarantee a suitable workspace for each child than their working-class counterparts 24 .

In the context of school closures, such disparities are likely to have important consequences for educational continuity. In line with this idea, a survey of approximately 4,000 parents in the United Kingdom confirmed that during lockdown, more than half of primary school children from the poorest families did not have access to their own study space and were less well equipped for distance learning than higher-income families 10 . Similarly, a survey of around 1,300 parents in the Netherlands found that during lockdown, children from working-class families had fewer computers at home and less room to study than upper/middle-class children 11 .

Data from non-Western countries highlight a more general digital divide, showing that developing countries have poorer access to digital equipment. For example, in India in 2018, only 10.7% of households possessed a digital device 25 , while in Pakistan in 2020, 31% of higher-education teachers did not have Internet access and 68.4% did not have a laptop 26 . In general, developing countries lack access to digital technologies 27 , 28 , and these difficulties of access are even greater in rural areas (for example, see ref. 29 ). Consequently, school closures have huge repercussions for the continuity of learning in these countries. For example, in India in 2018, only 11% of the rural and 40% of the urban population above 14 years old could use a computer and access the Internet 25 . Time spent on education during school closure decreased by 80% in Bangladesh 30 . A similar trend was observed in other countries 31 , with only 22% of children engaging in remote learning in Kenya 32 and 50% in Burkina Faso 33 . In Ghana, 26–32% of children spent no time at all on learning during the pandemic 34 . Beyond the overall digital divide, social class disparities are also evident in developing countries, with lower access to digital resources among households in which parental educational levels were low (versus households in which parental educational levels were high; for example, see ref. 35 for Nigeria and ref. 31 for Ecuador).

Unequal digital skills

In addition to unequal access to digital tools, there are also systematic variations in digital skills 36 , 37 (Fig. 1 ). Upper/middle-class families are more familiar with digital tools and resources and are therefore more likely to have the digital skills needed for distance learning 38 , 39 , 40 . These digital skills are particularly useful during school closures, both for students and for parents, for organizing, retrieving and correctly using the resources provided by the teachers (for example, sending or receiving documents by email, printing documents or using word processors).

Social class disparities in digital skills can be explained in part by the fact that children from upper/middle-class families have the opportunity to develop digital skills earlier than working-class families 41 . In member countries of the OECD (Organisation for Economic Co-operation and Development), only 23% of working-class children had started using a computer at the age of 6 years or earlier compared with 43% of upper/middle-class children 42 . Moreover, because working-class people tend to persist less than upper/middle-class people when confronted with digital difficulties 23 , the use of digital tools and resources for distance learning may interfere with the ability of parents to help children with their schoolwork.

Unequal use of digital tools

A third level of digital divide concerns variations in digital tool use 18 , 43 (Fig. 1 ). Upper/middle-class families are more likely to use digital resources for work and education 6 , 41 , 44 , whereas working-class families are more likely to use these resources for entertainment, such as electronic games or social media 6 , 45 . This divide is also observed among students, whereby working-class students tend to use digital technologies for leisure activities, whereas their upper/middle-class peers are more likely to use them for academic activities 46 and to consider that computers and the Internet provide an opportunity for education and training 23 . Furthermore, working-class families appear to regulate the digital practices of their children less 47 and are more likely to allow screens in the bedrooms of children and teenagers without setting limits on times or practices 48 .

In sum, inequalities in terms of digital resources, skills and use have strong implications for distance learning. This is because they make working-class students and parents particularly vulnerable when learning relies on extensive use of digital devices rather than on face-to-face interaction with teachers.

The cultural divide

Even if all three levels of digital divide were closed, upper/middle-class families would still be better prepared than working-class families to ensure educational continuity for their children. Upper/middle-class families are more familiar with the academic knowledge and skills that are expected and valued in educational settings, as well as with the independent, autonomous way of learning that is valued in the school culture and becomes even more important during school closure (Fig. 1 ).

Unequal familiarity with academic knowledge and skills

According to classical social reproduction theory 8 , 49 , school is not a neutral place in which all forms of language and knowledge are equally valued. Academic contexts expect and value culture-specific and taken-for-granted forms of knowledge, skills and ways of being, thinking and speaking that are more in tune with those developed through upper/middle-class socialization (that is, ‘cultural capital’ 8 , 50 , 51 , 52 , 53 ). For instance, academic contexts value interest in the arts, museums and literature 54 , 55 , a type of interest that is more likely to develop through socialization in upper/middle-class families than in working-class socialization 54 , 56 . Indeed, upper/middle-class parents are more likely than working-class parents to engage in activities that develop this cultural capital. For example, they possess more books and cultural objects at home, read more stories to their children and visit museums and libraries more often (for examples, see refs. 51 , 54 , 55 ). Upper/middle-class children are also more involved in extra-curricular activities (for example, playing a musical instrument) than working-class children 55 , 56 , 57 .

Beyond this implicit familiarization with the school curriculum, upper/middle-class parents more often organize educational activities that are explicitly designed to develop academic skills of their children 57 , 58 , 59 . For example, they are more likely to monitor and re-explain lessons or use games and textbooks to develop and reinforce academic skills (for example, labelling numbers, letters or colours 57 , 60 ). Upper/middle-class parents also provide higher levels of support and spend more time helping children with homework than working-class parents (for examples, see refs. 61 , 62 ). Thus, even if all parents are committed to the academic success of their children, working-class parents have fewer chances to provide the help that children need to complete homework 63 , and homework is more beneficial for children from upper-middle class families than for children from working-class families 64 , 65 .

School closures amplify the impact of cultural inequalities

The trends described above have been observed in ‘normal’ times when schools are open. School closures, by making learning rely more strongly on practices implemented at home (rather than at school), are likely to amplify the impact of these disparities. Consistent with this idea, research has shown that the social class achievement gap usually greatly widens during school breaks—a phenomenon described as ‘summer learning loss’ or ‘summer setback’ 66 , 67 , 68 . During holidays, the learning by children tends to decline, and this is particularly pronounced in children from working-class families. Consequently, the social class achievement gap grows more rapidly during the summer months than it does in the rest of the year. This phenomenon is partly explained by the fact that during the break from school, social class disparities in investment in activities that are beneficial for academic achievement (for example, reading, travelling to a foreign country or museum visits) are more pronounced.

Therefore, when they are out of school, children from upper/middle-class backgrounds may continue to develop academic skills unlike their working-class counterparts, who may stagnate or even regress. Research also indicates that learning loss during school breaks tends to be cumulative 66 . Thus, repeated episodes of school closure are likely to have profound consequences for the social class achievement gap. Consistent with the idea that school closures could lead to similar processes as those identified during summer breaks, a recent survey indicated that during the COVID-19 lockdown in the United Kingdom, children from upper/middle-class families spent more time on educational activities (5.8 h per day) than those from working-class families (4.5 h per day) 7 , 69 .

Unequal dispositions for autonomy and self-regulation

School closures have encouraged autonomous work among students. This ‘independent’ way of studying is compatible with the family socialization of upper/middle-class students, but does not match the interdependent norms more commonly associated with working-class contexts 9 . Upper/middle-class contexts tend to promote cultural norms of independence whereby individuals perceive themselves as autonomous actors, independent of other individuals and of the social context, able to pursue their own goals 70 . For example, upper/middle-class parents tend to invite children to express their interests, preferences and opinions during the various activities of everyday life 54 , 55 . Conversely, in working-class contexts characterized by low economic resources and where life is more uncertain, individuals tend to perceive themselves as interdependent, connected to others and members of social groups 53 , 70 , 71 . This interdependent self-construal fits less well with the independent culture of academic contexts. This cultural mismatch between interdependent self-construal common in working-class students and the independent norms of the educational institution has negative consequences for academic performance 9 .

Once again, the impact of these differences is likely to be amplified during school closures, when being able to work alone and autonomously is especially useful. The requirement to work alone is more likely to match the independent self-construal of upper/middle-class students than the interdependent self-construal of working-class students. In the case of working-class students, this mismatch is likely to increase their difficulties in working alone at home. Supporting our argument, recent research has shown that working-class students tend to underachieve in contexts where students work individually compared with contexts where students work with others 72 . Similarly, during school closures, high self-regulation skills (for example, setting goals, selecting appropriate learning strategies and maintaining motivation 73 ) are required to maintain study activities and are likely to be especially useful for using digital resources efficiently. Research has shown that students from working-class backgrounds typically develop their self-regulation skills to a lesser extent than those from upper/middle-class backgrounds 74 , 75 , 76 .

Interestingly, some authors have suggested that independent (versus interdependent) self-construal may also affect communication with teachers 77 . Indeed, in the context of distance learning, working-class families are less likely to respond to the communication of teachers because their ‘interdependent’ self leads them to respect hierarchies, and thus perceive teachers as an expert who ‘can be trusted to make the right decisions for learning’. Upper/middle class families, relying on ‘independent’ self-construal, are more inclined to seek individualized feedback, and therefore tend to participate to a greater extent in exchanges with teachers. Such cultural differences are important because they can also contribute to the difficulties encountered by working-class families.

The structural divide: unequal support from schools

The issues reviewed thus far all increase the vulnerability of children and students from underprivileged backgrounds when schools are closed. To offset these disadvantages, it might be expected that the school should increase its support by providing additional resources for working-class students. However, recent data suggest that differences in the material and human resources invested in providing educational support for children during periods of school closure were—paradoxically—in favour of upper/middle-class students (Fig. 1 ). In England, for example, upper/middle-class parents reported benefiting from online classes and video-conferencing with teachers more often than working-class parents 10 . Furthermore, active help from school (for example, online teaching, private tutoring or chats with teachers) occurred more frequently in the richest households (64% of the richest households declared having received help from school) than in the poorest households (47%). Another survey found that in the United Kingdom, upper/middle-class children were more likely to take online lessons every day (30%) than working-class students (16%) 12 . This substantial difference might be due, at least in part, to the fact that private schools are better equipped in terms of online platforms (60% of schools have at least one online platform) than state schools (37%, and 23% in the most deprived schools) and were more likely to organize daily online lessons. Similarly, in the United Kingdom, in schools with a high proportion of students eligible for free school meals, teachers were less inclined to broadcast an online lesson for their pupils 78 . Interestingly, 58% of teachers in the wealthiest areas reported having messaged their students or their students’ parents during lockdown compared with 47% in the most deprived schools. In addition, the probability of children receiving technical support from the school (for example, by providing pupils with laptops or other devices) is, surprisingly, higher in the most advantaged schools than in the most deprived 78 .

In addition to social class disparities, there has been less support from schools for African-American and Latinx students. During school closures in the United States, 40% of African-American students and 30% of Latinx students received no online teaching compared with 10% of white students 79 . Another source of inequality is that the probability of school closure was correlated with social class and race. In the United States, for example, school closures from September to December 2020 were more common in schools with a high proportion of racial/ethnic minority students, who experience homelessness and are eligible for free/discounted school meals 80 .

Similarly, access to educational resources and support was lower in poorer (compared with richer) countries 81 . In sub-Saharan Africa, during lockdown, 45% of children had no exposure at all to any type of remote learning. Of those who did, the medium was mostly radio, television or paper rather than digital. In African countries, at most 10% of children received some material through the Internet. In Latin America, 90% of children received some remote learning, but less than half of that was through the internet—the remainder being via radio and television 81 . In Ecuador, high-school students from the lowest wealth quartile had fewer remote-learning opportunities, such as Google class/Zoom, than students from the highest wealth quartile 31 .

Thus, the achievement gap and its accentuation during lockdown are due not only to the cultural and digital disadvantages of working-class families but also to unequal support from schools. This inequality in school support is not due to teachers being indifferent to or even supportive of social stratification. Rather, we believe that these effects are fundamentally structural. In many countries, schools located in upper/middle-class neighbourhoods have more money than those in the poorest neighbourhoods. Moreover, upper/middle-class parents invest more in the schools of their children than working-class parents (for example, see ref. 82 ), and schools have an interest in catering more for upper/middle-class families than for working-class families 83 . Additionally, the expectation of teachers may be lower for working-class children 84 . For example, they tend to estimate that working-class students invest less effort in learning than their upper/middle-class counterparts 85 . These differences in perception may have influenced the behaviour of teachers during school closure, such that teachers in privileged neighbourhoods provided more information to students because they expected more from them in term of effort and achievement. The fact that upper/middle-class parents are better able than working-class parents to comply with the expectations of teachers (for examples, see refs. 55 , 86 ) may have reinforced this phenomenon. These discrepancies echo data showing that working-class students tend to request less help in their schoolwork than upper/middle-class ones 87 , and they may even avoid asking for help because they believe that such requests could lead to reprimands 88 . During school closures, these students (and their families) may in consequence have been less likely to ask for help and resources. Jointly, these phenomena have resulted in upper/middle-class families receiving more support from schools during lockdown than their working-class counterparts.

Psychological effects of digital, cultural and structural divides

Despite being strongly influenced by social class, differences in academic achievement are often interpreted by parents, teachers and students as reflecting differences in ability 89 . As a result, upper/middle-class students are usually perceived—and perceive themselves—as smarter than working-class students, who are perceived—and perceive themselves—as less intelligent 90 , 91 , 92 or less able to succeed 93 . Working-class students also worry more about the fact that they might perform more poorly than upper/middle-class students 94 , 95 . These fears influence academic learning in important ways. In particular, they can consume cognitive resources when children and students work on academic tasks 96 , 97 . Self-efficacy also plays a key role in engaging in learning and perseverance in the face of difficulties 13 , 98 . In addition, working-class students are those for whom the fear of being outperformed by others is the most negatively related to academic performance 99 .

The fact that working-class children and students are less familiar with the tasks set by teachers, and less well equipped and supported, makes them more likely to experience feelings of incompetence (Fig. 1 ). Working-class parents are also more likely than their upper/middle-class counterparts to feel unable to help their children with schoolwork. Consistent with this, research has shown that both working-class students and parents have lower feelings of academic self-efficacy than their upper/middle-class counterparts 100 , 101 . These differences have been documented under ‘normal’ conditions but are likely to be exacerbated during distance learning. Recent surveys conducted during the school closures have confirmed that upper/middle-class families felt better able to support their children in distance learning than did working-class families 10 and that upper/middle-class parents helped their children more and felt more capable to do so 11 , 12 .

Pandemic disparity, future directions and recommendations

The research reviewed thus far suggests that children and their families are highly unequal with respect to digital access, skills and use. It also shows that upper/middle-class students are more likely to be supported in their homework (by their parents and teachers) than working-class students, and that upper/middle-class students and parents will probably feel better able than working-class ones to adapt to the context of distance learning. For all these reasons, we anticipate that as a result of school closures, the COVID-19 pandemic will substantially increase the social class achievement gap. Because school closures are a recent occurrence, it is too early to measure with precision their effects on the widening of the achievement gap. However, some recent data are consistent with this idea.

Evidence for a widening gap during the pandemic

Comparing academic achievement in 2020 with previous years provides an early indication of the effects of school closures during the pandemic. In France, for example, first and second graders take national evaluations at the beginning of the school year. Initial comparisons of the results for 2020 with those from previous years revealed that the gap between schools classified as ‘priority schools’ (those in low-income urban areas) and schools in higher-income neighbourhoods—a gap observed every year—was particularly pronounced in 2020 in both French and mathematics 102 .

Similarly, in the Netherlands, national assessments take place twice a year. In 2020, they took place both before and after school closures. A recent analysis compared progress during this period in 2020 in mathematics/arithmetic, spelling and reading comprehension for 7–11-year-old students within the same period in the three previous years 103 . Results indicated a general learning loss in 2020. More importantly, for the 8% of working-class children, the losses were 40% greater than they were for upper/middle-class children.

Similar results were observed in Belgium among students attending the final year of primary school. Compared with students from previous cohorts, students affected by school closures experienced a substantial decrease in their mathematics and language scores, with children from more disadvantaged backgrounds experiencing greater learning losses 104 . Likewise, oral reading assessments in more than 100 school districts in the United States showed that the development of this skill among children in second and third grade significantly slowed between Spring and Autumn 2020, but this slowdown was more pronounced in schools from lower-achieving districts 105 .

It is likely that school closures have also amplified racial disparities in learning and achievement. For example, in the United States, after the first lockdown, students of colour lost the equivalent of 3–5 months of learning, whereas white students were about 1–3 months behind. Moreover, in the Autumn, when some students started to return to classrooms, African-American and Latinx students were more likely to continue distance learning, despite being less likely to have access to the digital tools, Internet access and live contact with teachers 106 .

In some African countries (for example, Ethiopia, Kenya, Liberia, Tanzania and Uganda), the COVID-19 crisis has resulted in learning loss ranging from 6 months to more 1 year 107 , and this learning loss appears to be greater for working-class children (that is, those attending no-fee schools) than for upper/middle-class children 108 .

These findings show that school closures have exacerbated achievement gaps linked to social class and ethnicity. However, more research is needed to address the question of whether school closures differentially affect the learning of students from working- and upper/middle-class families.

Future directions

First, to assess the specific and unique impact of school closures on student learning, longitudinal research should compare student achievement at different times of the year, before, during and after school closures, as has been done to document the summer learning loss 66 , 109 . In the coming months, alternating periods of school closure and opening may occur, thereby presenting opportunities to do such research. This would also make it possible to examine whether the gap diminishes a few weeks after children return to in-school learning or whether, conversely, it increases with time because the foundations have not been sufficiently acquired to facilitate further learning 110 .

Second, the mechanisms underlying the increase in social class disparities during school closures should be examined. As discussed above, school closures result in situations for which students are unevenly prepared and supported. It would be appropriate to seek to quantify the contribution of each of the factors that might be responsible for accentuating the social class achievement gap. In particular, distinguishing between factors that are relatively ‘controllable’ (for example, resources made available to pupils) and those that are more difficult to control (for example, the self-efficacy of parents in supporting the schoolwork of their children) is essential to inform public policy and teaching practices.

Third, existing studies are based on general comparisons and very few provide insights into the actual practices that took place in families during school closure and how these practices affected the achievement gap. For example, research has documented that parents from working-class backgrounds are likely to find it more difficult to help their children to complete homework and to provide constructive feedback 63 , 111 , something that could in turn have a negative impact on the continuity of learning of their children. In addition, it seems reasonable to assume that during lockdown, parents from upper/middle-class backgrounds encouraged their children to engage in practices that, even if not explicitly requested by teachers, would be beneficial to learning (for example, creative activities or reading). Identifying the practices that best predict the maintenance or decline of educational achievement during school closures would help identify levers for intervention.

Finally, it would be interesting to investigate teaching practices during school closures. The lockdown in the spring of 2020 was sudden and unexpected. Within a few days, teachers had to find a way to compensate for the school closure, which led to highly variable practices. Some teachers posted schoolwork on platforms, others sent it by email, some set work on a weekly basis while others set it day by day. Some teachers also set up live sessions in large or small groups, providing remote meetings for questions and support. There have also been variations in the type of feedback given to students, notably through the monitoring and correcting of work. Future studies should examine in more detail what practices schools and teachers used to compensate for the school closures and their effects on widening, maintaining or even reducing the gap, as has been done for certain specific literacy programmes 112 as well as specific instruction topics (for example, ecology and evolution 113 ).

Practical recommendations

We are aware of the debate about whether social science research on COVID-19 is suitable for making policy decisions 114 , and we draw attention to the fact that some of our recommendations (Table 1 ) are based on evidence from experiments or interventions carried out pre-COVID while others are more speculative. In any case, we emphasize that these suggestions should be viewed with caution and be tested in future research. Some of our recommendations could be implemented in the event of new school closures, others only when schools re-open. We also acknowledge that while these recommendations are intended for parents and teachers, their implementation largely depends on the adoption of structural policies. Importantly, given all the issues discussed above, we emphasize the importance of prioritizing, wherever possible, in-person learning over remote learning 115 and where this is not possible, of implementing strong policies to support distance learning, especially for disadvantaged families.

Where face-to face teaching is not possible and teachers are responsible for implementing distance learning, it will be important to make them aware of the factors that can exacerbate inequalities during lockdown and to provide them with guidance about practices that would reduce these inequalities. Thus, there is an urgent need for interventions aimed at making teachers aware of the impact of the social class of children and families on the following factors: (1) access to, familiarity with and use of digital devices; (2) familiarity with academic knowledge and skills; and (3) preparedness to work autonomously. Increasing awareness of the material, cultural and psychological barriers that working-class children and families face during lockdown should increase the quality and quantity of the support provided by teachers and thereby positively affect the achievements of working-class students.

In addition to increasing the awareness of teachers of these barriers, teachers should be encouraged to adjust the way they communicate with working-class families due to differences in self-construal compared with upper/middle-class families 77 . For example, questions about family (rather than personal) well-being would be congruent with interdependent self-construals. This should contribute to better communication and help keep a better track of the progress of students during distance learning.

It is also necessary to help teachers to engage in practices that have a chance of reducing inequalities 53 , 116 . Particularly important is that teachers and schools ensure that homework can be done by all children, for example, by setting up organizations that would help children whose parents are not in a position to monitor or assist with the homework of their children. Options include homework help groups and tutoring by teachers after class. When schools are open, the growing tendency to set homework through digital media should be resisted as far as possible given the evidence we have reviewed above. Moreover, previous research has underscored the importance of homework feedback provided by teachers, which is positively related to the amount of homework completed and predictive of academic performance 117 . Where homework is web-based, it has also been shown that feedback on web-based homework enhances the learning of students 118 . It therefore seems reasonable to predict that the social class achievement gap will increase more slowly (or even remain constant or be reversed) in schools that establish individualized monitoring of students, by means of regular calls and feedback on homework, compared with schools where the support provided to pupils is more generic.

Given that learning during lockdown has increasingly taken place in family settings, we believe that interventions involving the family are also likely to be effective 119 , 120 , 121 . Simply providing families with suitable material equipment may be insufficient. Families should be given training in the efficient use of digital technology and pedagogical support. This would increase the self-efficacy of parents and students, with positive consequences for achievement. Ideally, such training would be delivered in person to avoid problems arising from the digital divide. Where this is not possible, individualized online tutoring should be provided. For example, studies conducted during the lockdown in Botswana and Italy have shown that individual online tutoring directly targeting either parents or students in middle school has a positive impact on the achievement of students, particularly for working-class students 122 , 123 .

Interventions targeting families should also address the psychological barriers faced by working-class families and children. Some interventions have already been designed and been shown to be effective in reducing the social class achievement gap, particularly in mathematics and language 124 , 125 , 126 . For example, research showed that an intervention designed to train low-income parents in how to support the mathematical development of their pre-kindergarten children (including classes and access to a library of kits to use at home) increased the quality of support provided by the parents, with a corresponding impact on the development of mathematical knowledge of their children. Such interventions should be particularly beneficial in the context of school closure.

Beyond its impact on academic performance and inequalities, the COVID-19 crisis has shaken the economies of countries around the world, casting millions of families around the world into poverty 127 , 128 , 129 . As noted earlier, there has been a marked increase in economic inequalities, bringing with it all the psychological and social problems that such inequalities create 130 , 131 , especially for people who live in scarcity 132 . The increase in educational inequalities is just one facet of the many difficulties that working-class families will encounter in the coming years, but it is one that could seriously limit the chances of their children escaping from poverty by reducing their opportunities for upward mobility. In this context, it should be a priority to concentrate resources on the most deprived students. A large proportion of the poorest households do not own a computer and do not have personal access to the Internet, which has important consequences for distance learning. During school closures, it is therefore imperative to provide such families with adequate equipment and Internet service, as was done in some countries in spring 2020. Even if the provision of such equipment is not in itself sufficient, it is a necessary condition for ensuring pedagogical continuity during lockdown.

Finally, after prolonged periods of school closure, many students may not have acquired the skills needed to pursue their education. A possible consequence would be an increase in the number of students for whom teachers recommend class repetitions. Class repetitions are contentious. On the one hand, class repetition more frequently affects working-class children and is not efficient in terms of learning improvement 133 . On the other hand, accepting lower standards of academic achievement or even suspending the practice of repeating a class could lead to pupils pursuing their education without mastering the key abilities needed at higher grades. This could create difficulties in subsequent years and, in this sense, be counterproductive. We therefore believe that the most appropriate way to limit the damage of the pandemic would be to help children catch up rather than allowing them to continue without mastering the necessary skills. As is being done in some countries, systematic remedial courses (for example, summer learning programmes) should be organized and financially supported following periods of school closure, with priority given to pupils from working-class families. Such interventions have genuine potential in that research has shown that participation in remedial summer programmes is effective in reducing learning loss during the summer break 134 , 135 , 136 . For example, in one study 137 , 438 students from high-poverty schools were offered a multiyear summer school programme that included various pedagogical and enrichment activities (for example, science investigation and music) and were compared with a ‘no-treatment’ control group. Students who participated in the summer programme progressed more than students in the control group. A meta-analysis 138 of 41 summer learning programmes (that is, classroom- and home-based summer interventions) involving children from kindergarten to grade 8 showed that these programmes had significantly larger benefits for children from working-class families. Although such measures are costly, the cost is small compared to the price of failing to fulfil the academic potential of many students simply because they were not born into upper/middle-class families.

The unprecedented nature of the current pandemic means that we lack strong data on what the school closure period is likely to produce in terms of learning deficits and the reproduction of social inequalities. However, the research discussed in this article suggests that there are good reasons to predict that this period of school closures will accelerate the reproduction of social inequalities in educational achievement.

By making school learning less dependent on teachers and more dependent on families and digital tools and resources, school closures are likely to greatly amplify social class inequalities. At a time when many countries are experiencing second, third or fourth waves of the pandemic, resulting in fresh periods of local or general lockdowns, systematic efforts to test these predictions are urgently needed along with steps to reduce the impact of school closures on the social class achievement gap.

Bambra, C., Riordan, R., Ford, J. & Matthews, F. The COVID-19 pandemic and health inequalities. J. Epidemiol. Commun. Health 74 , 964–968 (2020).

Google Scholar  

Johnson, P, Joyce, R & Platt, L. The IFS Deaton Review of Inequalities: A New Year’s Message (Institute for Fiscal Studies, 2021).

Education: from disruption to recovery. https://en.unesco.org/covid19/educationresponse (UNESCO, 2020).

Daszak, P. We are entering an era of pandemics—it will end only when we protect the rainforest. The Guardian (28 July 2020); https://www.theguardian.com/commentisfree/2020/jul/28/pandemic-era-rainforest-deforestation-exploitation-wildlife-disease

Dobson, A. P. et al. Ecology and economics for pandemic prevention. Science 369 , 379–381 (2020).

Article   CAS   PubMed   Google Scholar  

Harris, C., Straker, L. & Pollock, C. A socioeconomic related ‘digital divide’ exists in how, not if, young people use computers. PLoS ONE 12 , e0175011 (2017).

Article   PubMed   PubMed Central   Google Scholar  

Zhang, M. Internet use that reproduces educational inequalities: evidence from big data. Comput. Educ. 86 , 212–223 (2015).

Article   Google Scholar  

Bourdieu, P. & Passeron, J. C. Reproduction in Education, Society and Culture (Sage, 1990).

Stephens, N. M., Fryberg, S. A., Markus, H. R., Johnson, C. S. & Covarrubias, R. Unseen disadvantage: how American universities’ focus on independence undermines the academic performance of first-generation college students. J. Pers. Soc. Psychol. 102 , 1178–1197 (2012).

Article   PubMed   Google Scholar  

Andrew, A. et al. Inequalities in children’s experiences of home learning during the COVID-19 lockdown in England. Fisc. Stud. 41 , 653–683 (2020).

Bol, T. Inequality in homeschooling during the Corona crisis in the Netherlands. First results from the LISS Panel. Preprint at SocArXiv https://doi.org/10.31235/osf.io/hf32q (2020).

Cullinane, C. & Montacute, R. COVID-19 and Social Mobility. Impact Brief #1: School Shutdown (The Sutton Trust, 2020).

Bandura, A. Self-efficacy: toward a unifying theory of behavioral change. Psychol. Rev. 84 , 191–215 (1977).

Prior, D. D., Mazanov, J., Meacheam, D., Heaslip, G. & Hanson, J. Attitude, digital literacy and self efficacy: low-on effects for online learning behavior. Internet High. Educ. 29 , 91–97 (2016).

Robinson, L. et al. Digital inequalities 2.0: legacy inequalities in the information age. First Monday https://doi.org/10.5210/fm.v25i7.10842 (2020).

Cruz-Jesus, F., Vicente, M. R., Bacao, F. & Oliveira, T. The education-related digital divide: an analysis for the EU-28. Comput. Hum. Behav. 56 , 72–82 (2016).

Rice, R. E. & Haythornthwaite, C. In The Handbook of New Media (eds Lievrouw, L. A. & Livingstone S. M.), 92–113 (Sage, 2006).

Yates, S., Kirby, J. & Lockley, E. Digital media use: differences and inequalities in relation to class and age. Sociol. Res. Online 20 , 71–91 (2015).

Legleye, S. & Rolland, A. Une personne sur six n’utilise pas Internet, plus d’un usager sur trois manques de compétences numériques de base [One in six people do not use the Internet, more than one in three users lack basic digital skills] (INSEE Première, 2019).

Green, F. Schoolwork in lockdown: new evidence on the epidemic of educational poverty (LLAKES Centre, 2020); https://www.llakes.ac.uk/wp-content/uploads/2021/03/RP-67-Francis-Green-Research-Paper-combined-file.pdf

Vogels, E. Digital divide persists even as americans with lower incomes make gains in tech adoption (Pew Research Center, 2021); https://www.pewresearch.org/fact-tank/2021/06/22/digital-divide-persists-even-as-americans-with-lower-incomes-make-gains-in-tech-adoption/

McBurnie, C., Adam, T. & Kaye, T. Is there learning continuity during the COVID-19 pandemic? A synthesis of the emerging evidence. J. Learn. Develop. http://dspace.col.org/handle/11599/3720 (2020).

Baillet, J., Croutte, P. & Prieur, V. Baromètre du numérique 2019 [Digital barometer 2019] (Sourcing Crédoc, 2019).

Giraud, F., Bertrand, J., Court, M. & Nicaise, S. In Enfances de Classes. De l’inégalité Parmi les Enfants (ed. Lahire, B.) 933–952 (Seuil, 2019).

Ahamed, S. & Siddiqui, Z. Disparity in access to quality education and the digital divide (Ideas for India, 2020); https://www.ideasforindia.in/topics/macroeconomics/disparity-in-access-to-quality-education-and-the-digital-divide.html

Soomro, K. A., Kale, U., Curtis, R., Akcaoglu, M. & Bernstein, M. Digital divide among higher education faculty. Int. J. Educ. Tech. High. Ed. 17 , 21 (2020).

Meng, Q. & Li, M. New economy and ICT development in China. Inf. Econ. Policy 14 , 275–295 (2002).

Chinn, M. D. & Fairlie, R. W. The determinants of the global digital divide: a cross-country analysis of computer and internet penetration. Oxf. Econ. Pap. 59 , 16–44 (2006).

Lembani, R., Gunter, A., Breines, M. & Dalu, M. T. B. The same course, different access: the digital divide between urban and rural distance education students in South Africa. J. Geogr. High. Educ. 44 , 70–84 (2020).

Asadullah, N., Bhattacharjee, A., Tasnim, M. & Mumtahena, F. COVID-19, schooling, and learning (BRAC Institute of Governance & Development, 2020); https://bigd.bracu.ac.bd/wp-content/uploads/2020/06/COVID-19-Schooling-and-Learning_June-25-2020.pdf

Asanov, I., Flores, F., McKenzie, D., Mensmann, M. & Schulte, M. Remote-learning, time-use, and mental health of Ecuadorian high-school students during the COVID-19 quarantine. World Dev. 138 , 105225 (2021).

Kihui, N. Kenya: 80% of students missing virtual learning amid school closures—study. AllAfrica (18 May 2020); https://allafrica.com/stories/202005180774.html

Debenedetti, L., Hirji, S., Chabi, M. O. & Swigart, T. Prioritizing evidence-based responses in Burkina Faso to mitigate the economic effects of COVID-19: lessons from RECOVR (Innovations for Poverty Action, 2020); https://www.poverty-action.org/blog/prioritizing-evidence-based-responses-burkina-faso-mitigate-economic-effects-covid-19-lessons

Bosumtwi-Sam, C. & Kabay, S. Using data and evidence to inform school reopening in Ghana (Innovations for Poverty Action, 2020); https://www.poverty-action.org/blog/using-data-and-evidence-inform-school-reopening-ghana

Azubuike, O. B., Adegboye, O. & Quadri, H. Who gets to learn in a pandemic? Exploring the digital divide in remote learning during the COVID-19 pandemic in Nigeria. Int. J. Educ. Res. Open 2 , 100022 (2021).

Attewell, P. Comment: the first and second digital divides. Sociol. Educ. 74 , 252–259 (2001).

DiMaggio, P., Hargittai, E., Neuman, W. R. & Robinson, J. P. Social implications of the Internet. Annu. Rev. Sociol. 27 , 307–336 (2001).

Hargittai, E. Digital na(t)ives? Variation in Internet skills and uses among members of the ‘Net Generation’. Sociol. Inq. 80 , 92–113 (2010).

Iivari, N., Sharma, S. & Ventä-Olkkonen, L. Digital transformation of everyday life—how COVID-19 pandemic transformed the basic education of the young generation and why information management research should care? Int. J. Inform. Manag. 55 , 102183 (2020).

Wei, L. & Hindman, D. B. Does the digital divide matter more? Comparing the effects of new media and old media use on the education-based knowledge gap. Mass Commun. Soc. 14 , 216–235 (2011).

Octobre, S. & Berthomier, N. L’enfance des loisirs [The childhood of leisure]. Cult. Études 6 , 1–12 (2011).

Education at a glance 2015: OECD indicators (OECD, 2015); https://doi.org/10.1787/eag-2015-en

North, S., Snyder, I. & Bulfin, S. Digital tastes: social class and young people’s technology use. Inform. Commun. Soc. 11 , 895–911 (2008).

Robinson, L. & Schulz, J. Net time negotiations within the family. Inform. Commun. Soc. 16 , 542–560 (2013).

Bonfadelli, H. The Internet and knowledge gaps: a theoretical and empirical investigation. Eur. J. Commun. 17 , 65–84 (2002).

Drabowicz, T. Social theory of Internet use: corroboration or rejection among the digital natives? Correspondence analysis of adolescents in two societies. Comput. Educ. 105 , 57–67 (2017).

Nikken, P. & Jansz, J. Developing scales to measure parental mediation of young children’s Internet use. Learn. Media Technol. 39 , 250–266 (2014).

Danic, I., Fontar, B., Grimault-Leprince, A., Le Mentec, M. & David, O. Les espaces de construction des inégalités éducatives [The areas of construction of educational inequalities] (Presses Univ. de Rennes, 2019).

Goudeau, S. Comment l'école reproduit-elle les inégalités? [How does school reproduce inequalities?] (Univ. Grenoble Alpes Editions/Presses Univ. de Grenoble, 2020).

Bernstein, B. Class, Codes, and Control (Routledge, 1975).

Gaddis, S. M. The influence of habitus in the relationship between cultural capital and academic achievement. Soc. Sci. Res. 42 , 1–13 (2013).

Lamont, M. & Lareau, A. Cultural capital: allusions, gaps and glissandos in recent theoretical developments. Sociol. Theory 6 , 153–168 (1988).

Stephens, N. M., Markus, H. R. & Phillips, L. T. Social class culture cycles: how three gateway contexts shape selves and fuel inequality. Annu. Rev. Psychol. 65 , 611–634 (2014).

Lahire, B. Enfances de classe. De l’inégalité parmi les enfants [Social class childhood. Inequality among children] (Le Seuil, 2019).

Lareau, A. Unequal Childhoods: Class, Race, and Family Life (Univ. of California Press, 2003).

Bourdieu, P. La distinction. Critique sociale du jugement [Distinction: a social critique of the judgement of taste] (Éditions de Minuit, 1979).

Bradley, R. H., Corwyn, R. F., McAdoo, H. P. & Garcia Coll, C. The home environments of children in the United States part I: variations by age, ethnicity, and poverty status. Child Dev. 72 , 1844–1867 (2001).

Blevins‐Knabe, B. & Musun‐Miller, L. Number use at home by children and their parents and its relationship to early mathematical performance. Early Dev. Parent. 5 , 35–45 (1996).

LeFevre, J. A. et al. Pathays to mathematics: longitudinal predictors of performance. Child Dev. 81 , 1753–1767 (2010).

Lareau, A. Home Advantage. Social Class and Parental Intervention in Elementary Education (Falmer Press, 1989).

Guryan, J., Hurst, E. & Kearney, M. Parental education and parental time with children. J. Econ. Perspect. 22 , 23–46 (2008).

Hill, C. R. & Stafford, F. P. Allocation of time to preschool children and educational opportunity. J. Hum. Resour. 9 , 323–341 (1974).

Calarco, J. M. A Field Guide to Grad School: Uncovering the Hidden Curriculum (Princeton Univ. Press, 2020).

Daw, J. Parental income and the fruits of labor: variability in homework efficacy in secondary school. Res. Soc. Strat. Mobil. 30 , 246–264 (2012).

Rønning, M. Who benefits from homework assignments? Econ. Educ. Rev. 30 , 55–64 (2011).

Alexander, K. L., Entwisle, D. R. & Olson, L. S. Lasting consequences of the summer learning gap. Am. Sociol. Rev. 72 , 167–180 (2007).

Cooper, H., Nye, B., Charlton, K., Lindsay, J. & Greathouse, S. The effects of summer vacation on achievement test scores: a narrative and meta-analytic review. Rev. Educ. Res. 66 , 227–268 (1996).

Stewart, H., Watson, N. & Campbell, M. The cost of school holidays for children from low income families. Childhood 25 , 516–529 (2018).

Pensiero, N., Kelly, A. & Bokhove, C. Learning inequalities during the Covid-19 pandemic: how families cope with home-schooling (University of Southampton, 2020); https://doi.org/10.5258/SOTON/P0025

Stephens, N. M., Markus, H. R. & Townsend, S. S. Choice as an act of meaning: the case of social class. J. Pers. Soc. Psychol. 93 , 814–830 (2007).

Kraus, M. W., Piff, P. K. & Keltner, D. Social class, sense of control, and social explanation. J. Pers. Soc. Psychol. 97 , 992–1004 (2009).

Dittmann, A. G., Stephens, N. M. & Townsend, S. S. Achievement is not class-neutral: working together benefits pople from working-class contexts. J. Pers. Soc. Psychol. 119 , 517–539 (2020).

Zimmerman, B. J. Investigating self-regulation and motivation: historical background, methodological developments, and future prospects. Am. Educ. Res. J. 45 , 166–183 (2008).

Backer-Grøndahl, A., Nærde, A., Ulleberg, P. & Janson, H. Measuring effortful control using the children’s behavior questionnaire–very short form: modeling matters. J. Pers. Assess. 98 , 100–109 (2016).

Johnson, S. E., Richeson, J. A. & Finkel, E. J. Middle class and marginal? Socioeconomic status, stigma, and self-regulation at an elite university. J. Pers. Soc. Psychol. 100 , 838–852 (2011).

Størksen, I., Ellingsen, I. T., Wanless, S. B. & McClelland, M. M. The influence of parental socioeconomic background and gender on self-regulation among 5-year-old children in Norway. Early Educ. Dev. 26 , 663–684 (2015).

Brady, L. et al. 7 ways for teachers to truly connect with parents. Education Week (31 December 2020); https://www.edweek.org/leadership/opinion-7-ways-for-teachers-to-truly-connect-with-parents/2020/12

Montacute, R. Social mobility and Covid-19: implications of the Covid-19 crisis for educational inequality (Sutton Trust, 2020); https://dera.ioe.ac.uk/35323/2/COVID-19-and-Social-Mobility-1.pdf

Dorn, E., Hancock, B., Sarakatsannis, J. & Viruleg, E. COVID-19 and student learning in the United States: the hurt could last a lifetime (McKinsey & Company, 2020); https://www.mckinsey.com/industries/public-and-social-sector/our-insights/covid-19-and-student-learning-in-the-united-states-the-hurt-could-last-a-lifetime

Parolin, Z. & Lee, E. K. Large socio-economic, geographic and demographic disparities exist in exposure to school closures. Nat. Hum. Behav. 5 , 522–528 (2021).

Saavedra, J. A silent and unequal education crisis. And the seeds for its solution (World Bank, 2021); https://blogs.worldbank.org/education/silent-and-unequal-education-crisis-and-seeds-its-solution

Murray, B., Domina, T., Renzulli, L. & Boylan, R. Civil society goes to school: parent–teacher associations and the equality of educational opportunity. Russell Sage Found. J. Soc. Sci. 5 , 41–63 (2019).

Calarco, J. M. Avoiding us versus them: how schools’ dependence on privileged ‘helicopter’ parents influences enforcement of rules. Am. Sociol. Rev. 85 , 223–246 (2020).

Rist, R. Student social class and teacher expectations: the self-fulfilling prophecy in ghetto education. Harv. Educ. Rev. 40 , 411–451 (1970).

Tobisch, A. & Dresel, M. Negatively or positively biased? Dependencies of teachers’ judgments and expectations based on students’ ethnic and social backgrounds. Soc. Psychol. Educ. 20 , 731–752 (2017).

Brantlinger, E. Dividing Classes: How the Middle-class Negotiates and Rationalizes School Advantage (Routledge, 2003).

Calarco, J. M. ‘I need help!’ Social class and children’s help-seeking in elementary school. Am. Sociol. Rev. 76 , 862–882 (2011).

Calarco, J. M. The inconsistent curriculum: cultural tool kits and student interpretations of ambiguous expectations. Soc. Psychol. Quart. 77 , 185–209 (2014).

Goudeau, S. & Cimpian, A. How do young children explain differences in the classroom? Implications for achievement, motivation, and educational equity. Perspect. Psychol. Sci. 16 , 533–552 (2021).

Croizet, J. C., Goudeau, S., Marot, M. & Millet, M. How do educational contexts contribute to the social class achievement gap: documenting symbolic violence from a social psychological point of view. Curr. Opin. Psychol. 18 , 105–110 (2017).

Goudeau, S. & Croizet, J.-C. Hidden advantages and disadvantages of social class: how classroom settings reproduce social inequality by staging unfair comparison. Psychol. Sci. 28 , 162–170 (2017).

Kudrna, L., Furnham, A. & Swami, V. The influence of social class salience on self-assessed intelligence. Soc. Behav. Personal. 38 , 859–864 (2010).

Wiederkehr, V., Darnon, C., Chazal, S., Guimond, S. & Martinot, D. From social class to self-efficacy: internalization of low social status pupils’ school performance. Soc. Psychol. Educ. 18 , 769–784 (2015).

Jury, M., Smeding, A., Court, M. & Darnon, C. When first-generation students succeed at university: on the link between social class, academic performance, and performance-avoidance goals. Contemp. Educ. Psychol. 41 , 25–36 (2015).

Jury, M., Quiamzade, A., Darnon, C. & Mugny, G. Higher and lower status individuals’ performance goals: the role of hierarchy stability. Motiv. Sci. 5 , 52–65 (2019).

Autin, F. & Croizet, J.-C. Improving working memory efficiency by reframing metacognitive interpretation of task difficulty. J. Exp. Psychol. Gen. 141 , 610–618 (2012).

Schmader, T., Johns, M. & Forbes, C. An integrated process model of stereotype threat effects on performance. Psychol. Rev. 115 , 336–356 (2008).

Usher, E. L. & Pajares, F. Self-efficacy for self-regulated learning: a validation study. Educ. Psychol. Meas. 68 , 443–463 (2008).

Bruno, A., Jury, M., Toczek-Capelle, M.-C. & Darnon, C. Are performance-avoidance goals always deleterious for academic achievement in college? The moderating role of social class. Soc. Psychol. Educ. 22 , 539–555 (2019).

Holloway, S. D. et al. Parenting self-efficacy and parental involvement: mediators or moderators between socioeconomic status and children’s academic competence in Japan and Korea? Res. Hum. Dev. 13 , 258–272 (2016).

Tazouti, Y. & Jarlégan, A. The mediating effects of parental self-efficacy and parental involvement on the link between family socioeconomic status and children’s academic achievement. J. Fam. Stud. 25 , 250–266 (2019).

Andreu, S. et al. Évaluations 2020, repères CP, CE1: premiers résultats [2020 assessments, first and second grades benchmarks: first results] (Ministère de l’Éducation nationale, de la Jeunesse et des Sports, 2020); https://www.education.gouv.fr/evaluations-2020-reperes-cp-ce1-premiers-resultats-307122

Engzell, P., Frey, A. & Verhagen, M. D. Learning loss due to school closures during the COVID-19 pandemic. Proc. Natl Acad. Sci. USA 118 , e2022376118 (2021).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Maldonado, J. E. & De Witte, K. The effect of school closures on standardized student test outcomes (KU Leuven—Faculty of Economics and Business, 2020); https://limo.libis.be/primo-explore/fulldisplay?docid=LIRIAS3189074&context=L&vid=Lirias&search_scope=Lirias&tab=default_tab&lang=en_US

Domingue, B., Hough, H. J., Lang, D. & Yeatman, J. Changing patterns of growth in oral reading fluency during the COVID-19 pandemic (PACE, 2021); https://edpolicyinca.org/publications/changing-patterns-growth-oral-reading-fluency-during-covid-19-pandemic

Dorn, E., Hancock, B., Sarakatsannis, J. & Viruleg, E. COVID-19 and learning loss—disparities grow and students need help (McKinsey & Company, 2020); https://www.mckinsey.com/industries/public-and-social-sector/our-insights/covid-19-and-learning-loss-disparities-grow-and-students-need-help

Angrist, N. et al. Building back better to avert a learning catastrophe: estimating learning loss from COVID-19 school shutdowns in Africa and facilitating short-term and long-term learning recovery. Int. J. Educ. Dev. 84 , 102397 (2021).

Reddy, V., Soudien, C. & Winnaar, L. Disrupted learning during COVID-19: the impact of school closures on education outcomes in South Africa (The Conversation, 2020); https://theconversation.com/impact-of-school-closures-on-education-outcomes-in-south-africa-136889

Entwisle, D. R. & Alexander, K. L. Summer setback: race, poverty, school composition, and mathematics achievement in the first two years of school. Am. Sociol. Rev. 57 , 72–84 (1992).

Kieffer, M. J. Catching up or falling behind? Initial English proficiency, concentrated poverty, and the reading growth of language minority learners in the United States. J. Educ. Psychol. 100 , 851–868 (2008).

Calarco, J. M., Horn, I. & Chen, G. A. ‘You need to be more responsible’: how math homework operates as a status-reinforcing process in school. Preprint at SocArXiv https://doi.org/10.31235/osf.io/xf96q (2020).

Kaiper-Marquez, A. et al. On the fly: adapting quickly to emergency remote instruction in a family literacy program. Int. Rev. Educ. 66 , 1–23 (2020).

Barton, D. C. Impacts of the COVID‐19 pandemic on field instruction and remote teaching alternatives: results from a survey of instructors. Ecol. Evol. 10 , 12499–12507 (2020).

Article   PubMed Central   Google Scholar  

IJzerman, H. et al. Use caution when applying behavioural science to policy. Nat. Hum. Behav. 4 , 1092–1094 (2020).

Taylor, J. & Mallery, J. In person and online learning go together (Stanford Institute for Economic Policy Research, 2020); https://siepr.stanford.edu/research/publications/person-and-online-learning-go-together

Dietrichson, J., Bøg, M., Filges, T. & Klint Jørgensen, A. M. Academic interventions for elementary and middle school students with low socioeconomic status: a systematic review and meta-analysis. Rev. Educ. Res. 87 , 243–282 (2017).

Núñez, J. C. et al. Teachers’ feedback on homework, homework-related behaviors, and academic achievement. J. Educ. Res. 108 , 204–216 (2015).

Singh, R. et al. In Artificial Intelligence in Education (eds Biswas, G.et al.) 328–336 (Springer Berlin Heidelberg, 2011).

Harackiewicz, J. M., Rozek, C. S., Hulleman, C. S. & Hyde, J. S. Helping parents to motivate adolescents in mathematics and science: an experimental test of a utility-value intervention. Psychol. Sci. 23 , 899–906 (2012).

Jeynes, W. A meta-analysis of the efficacy of different types of parental involvement programs for urban students. Urban Educ. 47 , 706–742 (2012).

Mol, S. E., Bus, A. G., De Jong, M. T. & Smeets, D. J. Added value of dialogic parent–child book readings: a meta-analysis. Early Educ. Dev. 19 , 7–26 (2008).

Angrist, N., Bergman, P. & Matsheng, M. School’s out: experimental evidence on limiting learning loss using “low-tech” in a pandemic (National Bureau of Economic Research, 2021); https://www.nber.org/papers/w28205

Carlana, M. & La Ferrara, E. Apart but connected: online tutoring and student outcomes during the COVID-19 pandemic (Institute of Labor Economics, 2021); http://hdl.handle.net/10419/232846

Pagan, S. & Sénéchal, M. Involving parents in a summer book reading program to promote reading comprehension, fluency, and vocabulary in grade 3 and grade 5 children. Can. J. Educ. 37 , 1–31 (2014).

Sénéchal, M. & LeFevre, J. A. Parental involvement in the development of children’s reading skill: a five‐year longitudinal study. Child Dev. 73 , 445–460 (2002).

Starkey, P. & Klein, A. Fostering parental support for children’s mathematical development: an intervention with Head Start families. Early Educ. Dev. 11 , 659–680 (2000).

Buheji, M. et al. The extent of Covid-19 pandemic socio-economic impact on global poverty: a global integrative multidisciplinary review. Am. J. Econ. 10 , 213–224 (2020).

The world economy on a tightrope (OECD, 2020); http://www.oecd.org/economic-outlook/june-2020/

Martin, A., Markhvida, M., Hallegatte, S. & Walsh, B. Socio-economic impacts of COVID-19 on household consumption and poverty. Econ. Disasters Clim. Change 4 , 453–479 (2020).

Jetten, J., Mols, F. & Selvanathan, H. P. How economic inequality fuels the rise and persistence of the Yellow Vest movement. Int. Rev. Soc. Psychol. 33 , 2 (2020).

Wilkinson, R. G. & Pickett, K. E. Income inequality and social dysfunction. Annu. Rev. Sociol. 35 , 493–511 (2009).

Sommet, N., Morselli, D. & Spini, D. Income inequality affects the psychological health of only the people facing scarcity. Psychol. Sci. 29 , 1911–1921 (2018).

Hattie, J. Visible Learning: A Synthesis of over 800 Meta-analyses Relating to Achievement (Routledge, 2008).

Cooper, H., Charlton, K., Valentine, J. C., Muhlenbruck, L. & Borman, G. D. Making the most of summer school: a meta-analytic and narrative review. Monogr. Soc. Res. Child 65 , 1–127 (2000).

Heyns, B. Schooling and cognitive development: is there a season for learning? Child Dev. 58 , 1151–1160 (1987).

McCombs, J. S., Augustine, C. H. & Schwartz, H. L. Making Summer Count: How Summer Programs can Boost Children’s Learning (Rand Education, 2011).

Borman, G. D. & Dowling, N. M. Longitudinal achievement effects of multiyear summer school: evidence from the teach Baltimore randomized field trial. Educ. Eval. Policy 28 , 25–48 (2006).

Kim, J. S. & Quinn, D. M. The effects of summer reading on low-income children’s literacy achievement from kindergarten to grade 8: a meta-analysis of classroom and home interventions. Rev. Educ. Res. 83 , 386–431 (2013).

Download references

Acknowledgements

We thank G. Reis for editing the figure. The writing of this manuscript was supported by grant ANR-19-CE28-0007–PRESCHOOL from the French National Research Agency (S.G.).

Author information

Authors and affiliations.

Université de Poitiers, CNRS, CeRCA, Centre de Recherches sur la Cognition et l’Apprentissage, Poitiers, France

Sébastien Goudeau & Camille Sanrey

Université Clermont Auvergne, CNRS, LAPSCO, Laboratoire de Psychologie Sociale et Cognitive, Clermont-Ferrand, France

Arnaud Stanczak & Céline Darnon

School of Psychology, Cardiff University, Cardiff, UK

Antony Manstead

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Sébastien Goudeau .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Peer review information Nature Human Behaviour thanks Daniele Checchi and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Cite this article.

Goudeau, S., Sanrey, C., Stanczak, A. et al. Why lockdown and distance learning during the COVID-19 pandemic are likely to increase the social class achievement gap. Nat Hum Behav 5 , 1273–1281 (2021). https://doi.org/10.1038/s41562-021-01212-7

Download citation

Received : 15 March 2021

Accepted : 06 September 2021

Published : 27 September 2021

Issue Date : October 2021

DOI : https://doi.org/10.1038/s41562-021-01212-7

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Socioeconomic inequalities in psychosocial well-being among adolescents under the covid-19 pandemic: a cross-regional comparative analysis in hong kong, mainland china, and the netherlands.

• Gary Ka-Ki Chung
• Xiaoting Liu
• Roger Yat-Nork Chung

Social Psychiatry and Psychiatric Epidemiology (2024)

Digital gender gaps in Students’ knowledge, attitudes and skills: an integrative data analysis across 32 Countries

• Diego G. Campos
• Ronny Scherer

Education and Information Technologies (2024)

Microlearning as a Concept to Optimize Integrated Services for Racially/Ethnically Diverse Families of Autistic Children

• Zhiwen Xiao
• Sandra Vanegas

Journal of Developmental and Physical Disabilities (2024)

The ethicality of the COVID-19 response in children and adolescents

• Fiona McNicholas

Irish Journal of Medical Science (1971 -) (2024)

Health-related quality of life of young refugees in Germany during the COVID-19 pandemic: comparisons to non-refugees and pre-pandemic times

• Johanna Braig
• Pia Schmees
• Heike Eschenbeck

Current Psychology (2024)

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

The COVID-19 pandemic has changed education forever. This is how 

With schools shut across the world, millions of children have had to adapt to new types of learning. Image:  REUTERS/Gonzalo Fuentes

.chakra .wef-1c7l3mo{-webkit-transition:all 0.15s ease-out;transition:all 0.15s ease-out;cursor:pointer;-webkit-text-decoration:none;text-decoration:none;outline:none;color:inherit;}.chakra .wef-1c7l3mo:hover,.chakra .wef-1c7l3mo[data-hover]{-webkit-text-decoration:underline;text-decoration:underline;}.chakra .wef-1c7l3mo:focus,.chakra .wef-1c7l3mo[data-focus]{box-shadow:0 0 0 3px rgba(168,203,251,0.5);} Cathy Li

Farah lalani.

.chakra .wef-9dduvl{margin-top:16px;margin-bottom:16px;line-height:1.388;font-size:1.25rem;}@media screen and (min-width:56.5rem){.chakra .wef-9dduvl{font-size:1.125rem;}} Explore and monitor how .chakra .wef-15eoq1r{margin-top:16px;margin-bottom:16px;line-height:1.388;font-size:1.25rem;color:#F7DB5E;}@media screen and (min-width:56.5rem){.chakra .wef-15eoq1r{font-size:1.125rem;}} Education, Gender and Work is affecting economies, industries and global issues

.chakra .wef-1nk5u5d{margin-top:16px;margin-bottom:16px;line-height:1.388;color:#2846F8;font-size:1.25rem;}@media screen and (min-width:56.5rem){.chakra .wef-1nk5u5d{font-size:1.125rem;}} Get involved with our crowdsourced digital platform to deliver impact at scale

Stay up to date:, education, gender and work.

• The COVID-19 has resulted in schools shut all across the world. Globally, over 1.2 billion children are out of the classroom.
• As a result, education has changed dramatically, with the distinctive rise of e-learning, whereby teaching is undertaken remotely and on digital platforms.
• Research suggests that online learning has been shown to increase retention of information, and take less time, meaning the changes coronavirus have caused might be here to stay.

While countries are at different points in their COVID-19 infection rates, worldwide there are currently more than 1.2 billion children in 186 countries affected by school closures due to the pandemic. In Denmark, children up to the age of 11 are returning to nurseries and schools after initially closing on 12 March , but in South Korea students are responding to roll calls from their teachers online .

With this sudden shift away from the classroom in many parts of the globe, some are wondering whether the adoption of online learning will continue to persist post-pandemic, and how such a shift would impact the worldwide education market.

Even before COVID-19, there was already high growth and adoption in education technology, with global edtech investments reaching US$18.66 billion in 2019 and the overall market for online education projected to reach $350 Billion by 2025 . Whether it is language apps , virtual tutoring , video conferencing tools, or online learning software , there has been a significant surge in usage since COVID-19.

How is the education sector responding to COVID-19?

In response to significant demand, many online learning platforms are offering free access to their services, including platforms like BYJU’S , a Bangalore-based educational technology and online tutoring firm founded in 2011, which is now the world’s most highly valued edtech company . Since announcing free live classes on its Think and Learn app, BYJU’s has seen a 200% increase in the number of new students using its product, according to Mrinal Mohit, the company's Chief Operating Officer.

Tencent classroom, meanwhile, has been used extensively since mid-February after the Chinese government instructed a quarter of a billion full-time students to resume their studies through online platforms. This resulted in the largest “online movement” in the history of education with approximately 730,000 , or 81% of K-12 students, attending classes via the Tencent K-12 Online School in Wuhan.

Have you read?

The future of jobs report 2023, how to follow the growth summit 2023.

Other companies are bolstering capabilities to provide a one-stop shop for teachers and students. For example, Lark, a Singapore-based collaboration suite initially developed by ByteDance as an internal tool to meet its own exponential growth, began offering teachers and students unlimited video conferencing time, auto-translation capabilities, real-time co-editing of project work, and smart calendar scheduling, amongst other features. To do so quickly and in a time of crisis, Lark ramped up its global server infrastructure and engineering capabilities to ensure reliable connectivity.

Alibaba’s distance learning solution, DingTalk, had to prepare for a similar influx: “To support large-scale remote work, the platform tapped Alibaba Cloud to deploy more than 100,000 new cloud servers in just two hours last month – setting a new record for rapid capacity expansion,” according to DingTalk CEO, Chen Hang.

Some school districts are forming unique partnerships, like the one between The Los Angeles Unified School District and PBS SoCal/KCET to offer local educational broadcasts, with separate channels focused on different ages, and a range of digital options. Media organizations such as the BBC are also powering virtual learning; Bitesize Daily , launched on 20 April, is offering 14 weeks of curriculum-based learning for kids across the UK with celebrities like Manchester City footballer Sergio Aguero teaching some of the content.

What does this mean for the future of learning?

While some believe that the unplanned and rapid move to online learning – with no training, insufficient bandwidth, and little preparation – will result in a poor user experience that is unconducive to sustained growth, others believe that a new hybrid model of education will emerge, with significant benefits. “I believe that the integration of information technology in education will be further accelerated and that online education will eventually become an integral component of school education,“ says Wang Tao, Vice President of Tencent Cloud and Vice President of Tencent Education.

There have already been successful transitions amongst many universities. For example, Zhejiang University managed to get more than 5,000 courses online just two weeks into the transition using “DingTalk ZJU”. The Imperial College London started offering a course on the science of coronavirus, which is now the most enrolled class launched in 2020 on Coursera .

Many are already touting the benefits: Dr Amjad, a Professor at The University of Jordan who has been using Lark to teach his students says, “It has changed the way of teaching. It enables me to reach out to my students more efficiently and effectively through chat groups, video meetings, voting and also document sharing, especially during this pandemic. My students also find it is easier to communicate on Lark. I will stick to Lark even after coronavirus, I believe traditional offline learning and e-learning can go hand by hand."

These 3 charts show the global growth in online learning

The challenges of online learning.

There are, however, challenges to overcome. Some students without reliable internet access and/or technology struggle to participate in digital learning; this gap is seen across countries and between income brackets within countries. For example, whilst 95% of students in Switzerland, Norway, and Austria have a computer to use for their schoolwork, only 34% in Indonesia do, according to OECD data .

In the US, there is a significant gap between those from privileged and disadvantaged backgrounds: whilst virtually all 15-year-olds from a privileged background said they had a computer to work on, nearly 25% of those from disadvantaged backgrounds did not. While some schools and governments have been providing digital equipment to students in need, such as in New South Wales , Australia, many are still concerned that the pandemic will widenthe digital divide .

Is learning online as effective?

For those who do have access to the right technology, there is evidence that learning online can be more effective in a number of ways. Some research shows that on average, students retain 25-60% more material when learning online compared to only 8-10% in a classroom. This is mostly due to the students being able to learn faster online; e-learning requires 40-60% less time to learn than in a traditional classroom setting because students can learn at their own pace, going back and re-reading, skipping, or accelerating through concepts as they choose.

Nevertheless, the effectiveness of online learning varies amongst age groups. The general consensus on children, especially younger ones, is that a structured environment is required , because kids are more easily distracted. To get the full benefit of online learning, there needs to be a concerted effort to provide this structure and go beyond replicating a physical class/lecture through video capabilities, instead, using a range of collaboration tools and engagement methods that promote “inclusion, personalization and intelligence”, according to Dowson Tong, Senior Executive Vice President of Tencent and President of its Cloud and Smart Industries Group.

Since studies have shown that children extensively use their senses to learn, making learning fun and effective through use of technology is crucial, according to BYJU's Mrinal Mohit. “Over a period, we have observed that clever integration of games has demonstrated higher engagement and increased motivation towards learning especially among younger students, making them truly fall in love with learning”, he says.

A changing education imperative

It is clear that this pandemic has utterly disrupted an education system that many assert was already losing its relevance . In his book, 21 Lessons for the 21st Century , scholar Yuval Noah Harari outlines how schools continue to focus on traditional academic skills and rote learning , rather than on skills such as critical thinking and adaptability, which will be more important for success in the future. Could the move to online learning be the catalyst to create a new, more effective method of educating students? While some worry that the hasty nature of the transition online may have hindered this goal, others plan to make e-learning part of their ‘new normal’ after experiencing the benefits first-hand.

The importance of disseminating knowledge is highlighted through COVID-19

Major world events are often an inflection point for rapid innovation – a clear example is the rise of e-commerce post-SARS . While we have yet to see whether this will apply to e-learning post-COVID-19, it is one of the few sectors where investment has not dried up . What has been made clear through this pandemic is the importance of disseminating knowledge across borders, companies, and all parts of society. If online learning technology can play a role here, it is incumbent upon all of us to explore its full potential.

Our education system is losing relevance. Here's how to unleash its potential

3 ways the coronavirus pandemic could reshape education, celebrities are helping the uk's schoolchildren learn during lockdown, don't miss any update on this topic.

Create a free account and access your personalized content collection with our latest publications and analyses.

License and Republishing

World Economic Forum articles may be republished in accordance with the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Public License, and in accordance with our Terms of Use.

The views expressed in this article are those of the author alone and not the World Economic Forum.

Related topics:

The agenda .chakra .wef-n7bacu{margin-top:16px;margin-bottom:16px;line-height:1.388;font-weight:400;} weekly.

A weekly update of the most important issues driving the global agenda

.chakra .wef-1dtnjt5{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;-webkit-flex-wrap:wrap;-ms-flex-wrap:wrap;flex-wrap:wrap;} More on Health and Healthcare Systems .chakra .wef-17xejub{-webkit-flex:1;-ms-flex:1;flex:1;justify-self:stretch;-webkit-align-self:stretch;-ms-flex-item-align:stretch;align-self:stretch;} .chakra .wef-nr1rr4{display:-webkit-inline-box;display:-webkit-inline-flex;display:-ms-inline-flexbox;display:inline-flex;white-space:normal;vertical-align:middle;text-transform:uppercase;font-size:0.75rem;border-radius:0.25rem;font-weight:700;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;line-height:1.2;-webkit-letter-spacing:1.25px;-moz-letter-spacing:1.25px;-ms-letter-spacing:1.25px;letter-spacing:1.25px;background:none;padding:0px;color:#B3B3B3;-webkit-box-decoration-break:clone;box-decoration-break:clone;-webkit-box-decoration-break:clone;}@media screen and (min-width:37.5rem){.chakra .wef-nr1rr4{font-size:0.875rem;}}@media screen and (min-width:56.5rem){.chakra .wef-nr1rr4{font-size:1rem;}} See all

How midwife mentors are making it safer for women to give birth in remote, fragile areas

Anna Cecilia Frellsen

May 9, 2024

From Athens to Dhaka: how chief heat officers are battling the heat

Angeli Mehta

May 8, 2024

How a pair of reading glasses could increase your income

Emma Charlton

Nigeria is rolling out Men5CV, a ‘revolutionary’ meningitis vaccine

5 conditions that highlight the women’s health gap

Kate Whiting

May 3, 2024

How philanthropy is empowering India's mental health sector

Kiran Mazumdar-Shaw

May 2, 2024

• International edition
• Australia edition
• Europe edition

The future of online learning: the long-term trends accelerated by Covid-19

With the technology now available, it’s clear that simply broadcasting pre-recorded lectures is no longer an option for forward-thinking universities

For Prof John Domingue, director of the Open University’s pioneering research and development lab, the Knowledge Media Institute (KMI), the “online genie” is out of the bottle and won’t go back in.

“It’s slightly galling to see some universities trying to replicate online almost exactly what they delivered face-to-face before Covid. Standing before a camera and broadcasting is not online teaching. You need to do things differently,” he says.

So what can universities undertake to make online learning more than just a heavy focus on streaming and recording technology? Domingue points to artificial intelligence (AI) and the concept of an online library for educators based on a Google search engine dedicated to education, and a Netflix-style recommendation tool that tracks down content to suit a lecturer’s own field, based on previous searches.

KMI is currently developing a personalised AI assistant or chatbot, an AI career coach and other tools that can analyse essays for marking and set up quizzes on revision topics.

Personalisation is also key to giving students and lecturers a better online experience. In 2017, Oxford’s Saïd Business School installed the first immersive virtual classroom of its kind in the UK: a bank of 27 HD screens able to simultaneously support up to 84 students from across the globe, called the Oxford Hub for International Virtual Education (or HIVE). An in-room camera follows lecturers moving around the room, who can respond – as in real life – to visual cues from and talk directly to individual students.

While such technology could be prohibitively expensive for many institutions, Duncan Peberdy, a consultant specialising in tech-enabled learning spaces and former adviser at the educational IT body, Jisc, says a much cheaper alternative in the form of a 3-4m wide screen offering a different dynamic based on simplified specifications has been developed by ViewSonic. “We are now in talks with two UK universities to jointly develop it on their campuses,” he says.

Meanwhile on UK campuses, many universities are striving to make the online experience more than just a lecturer broadcasting in front of a camera.

“We didn’t want that approach so we ‘shifted’ academics who were simply recurating their material with PowerPoint slides and brought in new hardware and specialists to assist them,” says Guy Daly, deputy vice-chancellor (education and students) at Coventry University.

“We realised our academics either needed the skills or support to deliver online learning in a very engaging way in a now very different world. Since March, we’ve repurposed 2,500 course modules at under- and postgraduate level for delivery in the first term of this academic year.”

Coventry has moved virtually all its student assessments and exams online. “We also used to talk about the death of the traditional lecture and bringing in more student activity-based learning as opposed to traditional didactic methods, but we’ve accelerated that journey due to Covid,” says Daly.

Wholesale and now permanent changes have gone hand in hand with the launches of Coventry’s first online postgraduate certificate in education and the first online nursing degree in England.

Many taught postgrad students, particularly those using labs, have been among students hardest hit, according to Prof Danielle George, associate dean for teaching and learning at the University of Manchester. “They only have one year to ensure they receive all their intended learning outcomes from their course. So we’ve invested in software to enable them to do prep work at home so they will then need less time in the lab itself,” she says.

“We have also helped them with time management, which is absolutely key [during short courses]. Covid took away their daily structure of going from room to room on campus so we’ve timetabled asynchronous activities – their lecturer will, say, be available ‘live’ at 9am to deliver a lecture and then answer questions, or they can choose to watch a recorded version later in their own time.

“My best advice to postgrads is to get involved in anything to do with induction – we’ve invested a lot more energy, time and passion in this area than we’ve done before and put on numerous practical online sessions,” says George.

• Postgraduates
• Higher education

Most viewed

• Reference Manager
• Simple TEXT file

People also looked at

Original research article, impact of the covid-19 pandemic on online learning in higher education: a bibliometric analysis.

• 1 Faculty of Public Administration, University of Ljubljana, Ljubljana, Slovenia
• 2 Department of Primary Level Education, University of the Aegean, Rhodes, Greece

The outbreak of the COVID-19 pandemic significantly disrupted higher education by forcing the transition to online learning, which became a mandatory teaching process during the lockdowns. Although the epidemiological situation has gradually improved since then, online learning is becoming ever more popular as it provides new learning opportunities. Therefore, the paper aims to present recent research trends concerning online learning in higher education during the COVID-19 pandemic by using selected bibliometric approaches. The bibliometric analysis is based on 8,303 documents from the Scopus database published between January 2020 and March 2022, when repeated lockdowns meant most countries were experiencing constant disruptions to the educational process. The results show that the COVID-19 pandemic increased interest in online learning research, notably in English-speaking and Asian countries, with most research being published in open-access scientific journals. Moreover, the topics most frequently discussed in the online learning research during the COVID-19 pandemic were ICT and pedagogy, technology-enhanced education, mental health and well-being, student experience and curriculum and professional development. Finally, the COVID-19 pandemic encouraged explorations of emergency remote learning approaches like e-learning, distance learning and virtual learning, which are intended to limit physical contact between teachers and students, where the specific requirements of a given field of study often guide which online learning approach is the most suitable. The findings add to the existing body of scientific knowledge and support the evidence-based policymaking needed to ensure sustainable higher education in the future.

1. Introduction

The outbreak of the COVID-19 pandemic significantly disrupted higher education by forcing the transition to online learning, which became a mandatory teaching process during the lockdowns ( Aristovnik et al., 2020a ). Despite the educational process saw disruptions on all levels of education, i.e., primary, secondary and tertiary ( Tang, 2023 ), as well as in adult education ( James and Thériault, 2020 ), worker education ( Dedeilia et al., 2023 ) and lifelong education ( Waller et al., 2020 ), higher education students proved to be one of the worst affected groups because the social distancing measures, on top of their education, challenged their financial and housing situation ( Aristovnik et al., 2020a ). Challenges arising from the density of students in educational facilities (e.g., campuses, faculties, dormitories etc.) meant higher education institutions were forced to offer education relying on various information and communication technologies (ICTs) and tried to ensure education comparable in quality to traditional learning, noting that the quality of online learning delivery holds important implications for student satisfaction and student performance ( Keržič et al., 2021 ). Nevertheless, the lockdown periods were devastating for many students also in terms of their emotional functioning ( Raccanello et al., 2022 ). The COVID-19 pandemic eventually grew more predictable and manageable, allowing higher education institutions to gradually shift back to traditional learning approaches. Although the epidemiological situation has improved over time, online learning is becoming increasingly popular as it provides new learning opportunities, especially when combined with traditional learning.

The rapid, yet from the health protection point of view necessary ( Aristovnik et al., 2020b ), shift from traditional learning to online learning considerably affected teaching and learning. The transition to online learning was made without adequate consideration of whether the study materials and teaching methods were suitable for this mode of higher education delivery. This was an ad hoc shift in a situation of great uncertainty for both teachers and students. The transition to online learning has also brought to the surface gaps in higher education providers’ preparedness and their lack of ICT infrastructure, resulting in unequal access to quality education for all, particularly students from rural areas and regions with lower socio-economic development. It is important to note here that the rapid shift to an online learning environment in emergency circumstances should not be confused with properly planned online education equipped with appropriate infrastructure that enables and supports pedagogical work and study in an online environment ( Hodges et al., 2020 ; Fuchs, 2022 ; Misiejuk et al., 2023 ). Apart from the changes in teaching and learning, the social aspect of students’ lives has been affected as well. The most worrying consequence has been social isolation leading to a lack of crucial social interaction for students ( Elmer et al., 2020 ; Bonsaksen et al., 2021 ; Fried et al., 2021 ; Van der Graaf et al., 2021 ) and in some cases also in coronavirus-related post-traumatic stress syndrome (PTSD) ( Ochnik et al., 2021 ). According to Gavriluţă et al. (2022) , three dimensions affected students during the COVID-19 pandemic: educational, social, and emotional. The transition from traditional to online learning entailed a significant transformation in education, requiring changes in teaching practices and new learning approaches. Further, the social aspect of the COVID-19 pandemic and associated lockdowns is evident in the absence of relational, economic and professional problems (in)directly affecting the transition to adulthood. The new reality changed attitudes to various aspects of life and, in turn, also affected emotional responsiveness. Briefly, substantial changes to everyday student lives were made during the COVID-19 pandemic that may hold far-reaching effects of currently unknown scope in the near and distant future ( Campos et al., 2022 ; Gao et al., 2022 ; Keržič et al., 2022 ; Rasli et al., 2022 ).

Therefore, the educational community requires greater insights into different aspects of the COVID-19 pandemic’s impact on online learning, e.g., students, teachers, pedagogy, ICT technology, online learning approaches and implications for various fields of study. In the context of higher education, some bibliometric studies (e.g., Gurcan et al., 2022 ; Saqr et al., 2023 ) have already sought to address issues involving online learning during the pandemic. Yet, they relied on a limited and narrow bibliographic dataset of peer-reviewed literature or lacked a qualitative synthesis of the results beyond the metrics, thereby neglecting some general comprehensive outlines of the global research into the topic ( Saqr et al., 2023 ). Moreover, despite some bibliometric studies focusing on technical aspects (e.g., Navarro-Espinosa et al., 2021 ; Bozkurt, 2022 ; Tlili et al., 2022 ), the identification of the most effective ICT tools for specific online learning approaches remains unclear. Finally, there are also some bibliometric studies that attempt to determine the effectiveness of online learning in providing higher education ( Brika et al., 2021 ; Baber et al., 2022 ; Bilal et al., 2022 ; Bozkurt, 2022 ; Fauzi, 2022 ; Küçük-Avci et al., 2022 ; Yan et al., 2022 ), however, they often overlook the specific requirements of individual fields of study, thereby neglecting the crucial aspect of tailoring online learning provision to different disciplines.

The bibliometric study presented in the paper accordingly aims to fill the presented gaps in the literature. Specifically, it aims to present a global overview of the recent research trends in online learning in higher education using a comprehensive dataset of literature encompassing different varieties of online learning approaches that can facilitate online learning during the COVID-19 pandemic, provide some relevant qualitative synthesis of the results beyond the metrics and examine the relationships between ICT tools, online learning approaches and fields of study. Thus, the present bibliometric study, focusing on higher education, tries to answer the following three research questions:

• RQ1: What is the current state of the online learning research by conducting a descriptive overview and identifying top-cited documents?

• RQ2: What is the scientific production of online learning research across countries and sources?

• RQ3: Which are the main research hotspots and concepts in online learning research?

The remainder of the paper is structured as follows. The next section provides a literature review of recent bibliometric studies. The following section outlines the materials and methods applied in the study before the results of the present bibliometric analysis are described in the next section. At the end, the final section provides a discussion and conclusion while summarizing the main findings and implications.

2. Literature review

The outbreak of the COVID-19 pandemic led many governments to expand the use of online learning approaches as a solution to the global health challenge. Researchers thus showed rising interest in investigating the field of online learning, its dimensions, and its trends on all levels of education, particularly higher education. Such research relied heavily on bibliometric approaches to analyzing scientific research in the higher education context. Pham et al. (2022) concluded based on the 414 articles that although in the decades prior, there was an increase in the number of articles touching on the components of e-learning, such as the learning management system, this rise was accelerated during the pandemic in both developed and developing countries. This may be attributed to the attention of governmental policies that considered the topic of e-learning to be critical and worthy of priority. Similarly, Fauzi (2022) investigated 1,496 articles and concluded that the research focused on a few specific topics. The first is the delivery factor, which refers to selecting the appropriate learning practices. The second is the health and safety factor that relates to minimizing any risk that e-learning could bring to the mental and physical health of learners or teachers, such as stress, anxiety or even depression. The third topic refers to the field of study and the impact of e-learning. In areas like medical education, where clinical activities and labs have to be attended in person, some online learning approaches might be less appropriate than when used in other areas, such as social studies, where the requirements are less complex or different. Zhang et al. (2022) confirmed this finding after performing bibliometric research on 1,061 articles published between January 2020 and August 2021. They explained that theorists and researchers showed a growing interest in ways to respond to crises, such as the pandemic, and how to develop the best practices to ensure the quality and efficiency of e-learning. Examples of such practices might be inquiry-oriented learning and hands-on activities. This could derive from the already existing tendency of education researchers to respond to unprecedented global challenges or changes. The authors explain that this conclusion addresses interest in e-learning practices holistically.

In the same context, Yan et al. (2022) employed a bibliometric approach and identified that various digital tools are used in e-learning in the field of health studies. After investigating 132 studies, they concluded that selecting appropriate tools depends on many factors, including the field of a given course, the aims, and their effectiveness. They add that these findings can be significant for groups of people such as experts or trainee teachers. Okoro et al. (2022) researched 1,722 articles published between 2012 and 2021 and detected a surge in interest in the mental health of postgraduate students, as revealed by the research trends discussed in these articles. Still, they describe this surge as having been greater between 2020 and 2021, which may be attributed to the COVID-19 restrictions and their implications. Moreover, they believe that this research focus will likely continue soon.

After looking at 2,307 articles published between 2017 and 2021, Baber et al. (2022) detected an increasing trend in researching digital literacy. While this was underway before the pandemic, the latter caused a statistically significant further surge. Digital literacy is approached in the studied articles through parameters like instruction, teachers, learners, ICT and its applications, content knowledge, competencies, skills, perceptions, and higher education. It is also associated with acquiring the qualities required to deal with topics such as misinformation, fake news, technological content knowledge, health literacy, COVID-19, and distance education. The authors state that their study identified dynamics hidden in these research trends, which will likely continue in the next few years.

In higher education specifically, based on 602 articles, Brika et al. (2021) corroborated the growing trend of publishing articles on e-learning during the pandemic and outlined certain sub-topics of it, namely: motivation and students’ attitudes; blended and virtual learning comparison; types of online assessment; stress, anxiety and mental health; strategies to improve learners’ skills; quality; performance of the education delivered; challenges; and the potential of technology to lead to change and reform of higher education syllabi or curricula. The scope of those articles was to paint a bigger picture of how higher education communities and institutions use and treat online learning. This is expected to help with efficient decision-making in the future in order to have better results and functions in higher education and appropriate response to crises.

The bibliometric studies carried out during the pandemic identified a trend among researchers in higher education institutions to investigate more the technology factor and how the progress of the Internet, along with information and communication technologies generally, can further assist new modes of learning, such as online learning and distance learning. This might be attributed to a vision for a better means for new types of learning, as Küçük-Avci et al. (2022) claimed after carrying out a bibliometric analysis of 1,547 articles published between 2020 and 2021. The authors detected certain trends regarding distance learning in higher education. A main finding of their study, along with the increase in studies on distance education and e-learning in higher education, is that before the pandemic, the fact that these approaches were not so mandatory meant there was greater efficiency, probably due to the learners’ motivation. The authors further claim that researchers show a stronger interest in the technological means that can assist these types of learning. In addition, while researching 1,986 articles, Bozkurt (2022) established an increase in the implementation of blended learning by researchers who also aim to investigate the relationship between technological applications and learning institutions. Within these tendencies, researchers consider four thematic fields: a comparison of online and onsite learning with regard to effectiveness and efficiency; the experience, impressions and attitudes of stakeholders and learning community members with respect to blended learning; teacher training and curriculum development that will assure the appropriate and challenge-free implementation of blended learning; and the use of mostly a quantitative approach to research of blended learning.

Bilal et al. (2022) also examined research trends concerned with e-learning in higher education during the COVID-19 period by researching 1,595 studies published between 2020 and 2021. The four main trends they identified were supplementary to those mentioned by other authors: the first is about the challenges regarding online learning or blended learning along with the appropriate strategies in response; the second is student-centered collaborative learning and appropriate curriculum design; the third concerns home-based learning through a type of laboratory and the general conditions surrounding it; and the fourth addresses teachers’ background, training, professional competencies and interdisciplinary learning.

Tlili et al. (2022) focused on mapping COVID-19’s impact on Massive Open Online Courses (MOOCs). The overall finding from the 108 articles they considered is that there has been growing interest in these courses generally, and more specifically in research around their function and quality. This interest encompasses the main features of such courses, which provide easy accessibility and flexibility. However, they noted that this interest followed another trend among researchers in the context. In other words, the countries that published on MOOCs before the pandemic are the same countries that published during the period under study. Moreover, they stated that there is interest in the technical characteristics and requirements of such courses. Finally, the authors concluded that although most MOOCs were ICT courses, research has escalated into courses that refer to business, personal development or the humanities.

Several conclusions can be drawn from the above bibliometric studies. First, the series of bibliometric studies conducted during the pandemic demonstrates the rise of interest in online learning in higher education during the pandemic. Of course, there was a tendency toward e-learning before the pandemic, but between 2020 and 2022, this seems to have accelerated. The phenomenon is more intense in countries such as the USA, Canada, Australia, the UK, India and China. Concerning the area of study, the focus of researchers appears to be greater in fields such as Engineering, Sciences, and Health Sciences, albeit all fields seem to be investigated ( Djeki et al., 2022 ; Pham et al., 2022 ; Vaicondam et al., 2022 ; Zhang et al., 2022 ). Various studies have focused on determining the effectiveness of e-learning classes and courses or pointing out parameters that influence their effectiveness. These could be the appropriate conditions or subtopics like motivation, blended learning, learning tools, teacher training, cooperation between different institutions or efficient practices ( Brika et al., 2021 ; Baber et al., 2022 ; Bilal et al., 2022 ; Bozkurt, 2022 ; Fauzi, 2022 ; Küçük-Avci et al., 2022 ; Yan et al., 2022 ). A specific trend of authors is to examine virtual classes and laboratories ( Kartimi et al., 2022 ; Rojas-Sánchez et al., 2022 ; Zhang et al., 2022 ). Finally, there is a focus on the technology factor. Namely, researchers have concentrated on technical issues and conditions related to e-learning courses and their proper functioning ( Navarro-Espinosa et al., 2021 ; Bozkurt, 2022 ; Tlili et al., 2022 ).

3. Materials and methods

Comprehensive bibliometric data on online learning research during the COVID-19 pandemic were retrieved on 1 March 2022 from Scopus, a world-leading bibliographic database of peer-reviewed literature. The Scopus database was preferred because it has a broader coverage of scientific research than other databases such as Web of Science ( Falagas et al., 2008 ). This was confirmed by an initial search using the same search query in each database, revealing that Scopus provided more relevant documents than Web of Science. Moreover, compared to the Scopus database, the Web of Science has been found to be a database that significantly underrepresents the scientific disciplines of the Social Sciences and the Arts and Humanities ( Mongeon and Paul-Hus, 2016 ). Although English dominates in both Scopus and Web of Science, Scopus generally offers wider coverage of non-English documents, given that the titles, abstracts, and keywords are in English ( Vera-Baceta et al., 2019 ). According to the basic statistical theory, which can also be applied in the context of bibliometric analysis, larger samples lead to analytical outcomes that are likely to be more accurate ( Rogers et al., 2020 ). Therefore, Scopus appears to be a more relevant bibliographic database meeting the specifics of online learning research during the COVID-19 pandemic.

The search strategy was based on title, abstract, and keywords search using the advanced search engine and the search query covered keywords related to different online learning types (using the Boolean operator ‘OR’) and the COVID-19 pandemic (using the Boolean operator ‘AND’). The search was further limited to the period 2020–2022 (using the Boolean operator ‘AND’) to capture documents published between January 2020 and March 2022, when most countries were experiencing constant disruptions in the educational process imposed by repeated lockdowns. As the search query had no language restrictions, the full text of the obtained documents can be in any language, provided that the titles, abstracts, and keywords are in English. Therefore, the language has no impact on the results, as the bibliometric analysis is conducted solely based on the titles, abstracts, and keywords of the documents. According to the presented search query, 9,921 documents were obtained. After further revising the obtained documents, it was identified that some of them are not explicitly related to the context of higher education. By machine screening of documents by title, abstract, and keywords, those related to lower levels of education (i.e., primary and secondary education), as well as adult and worker education (i.e., lifelong education), were excluded from the database. There were 1,618 or 16% of such documents. The remaining 8,303 documents were identified as eligible for further bibliometric examination of online learning research during the COVID-19 pandemic. The bibliometric analysis utilized several bibliometric approaches ( Figure 1 ).

Figure 1 . Bibliometric approaches used in the bibliometric analysis. Own elaboration.

First, a descriptive overview was conducted to examine particular general bibliometric items, including timespan, number of (all, cited, single-authored) documents, authors, sources and author keywords and authors, references, and citations per document as well as to identify the most relevant documents. Scientific production was also examined to determine the most relevant countries and sources. Finally, network analysis was performed to identify the research hotspots according to the keyword co-occurrence network and examine the relationship between the main concepts based on a three-field plot analysis. The presented bibliometric approaches required the use of several different software tools. The descriptive overview was conducted using the Python Data Analysis Library Pandas ( McKinney, 2012 ), scientific production was visualized by the Python Visualization Library Matplotlib ( Hunter, 2007 ), while network analysis was performed using VOSviewer (keyword co-occurrence) ( Van Eck and Waltman, 2010 ) and the Python Visualization Library Plotly (a three-field plot) ( Pandey and Panchal, 2020 ). Specifically, the calculation for the three-field plot analysis included the following steps. Suppose that C 1 , C 2 , … , C m are analysed concepts where each concept C i is defined by a set of keywords and represented by binary indicators W i 1 , W i 2 , … , W i k i , expressed as C i = max j = 1 , … , k i W i j for i = 1 , … , m (matrix column). Using this notation, the relationship between C i and C j can be defined as C 1 T ∗ C j (matrix multiplication) where i and j are from three different sets (ICT tools, online learning approaches, fields of study).

The descriptive overview presented in Table 1 shows the main characteristics of online learning and COVID-19 research in the higher education context. This research area covers a total of 8,303 documents (of which 7,922 (95%) have the full text in English) published in 2,447 sources between January 2020 and March 2022. Slightly less than half (46%) of these documents have at least one citation, while a relatively small number (15%) were written by a single author. The average number of references per document in this research area is 31.39, which is below the general scientific area of Educational Research (44.00) ( Patience et al., 2017 ), suggesting that online learning research during the COVID-19 pandemic is grounded on fewer existing studies than general research. Finally, 3.50 citations per document can be observed for this research area. Due to the potential benefits of online learning, especially when combined with the traditional learning approaches and hence the development of the blended learning environment, this research is expected to further develop and be extended in the ensuing years ( Fauzi, 2022 ). Further, upon analyzing the documents, it is evident that the average year of references is 2014.03, with an h-index of 60 (indicating at least 60 papers with 60 or more citations each) and a g-index of 94 (denoting that the top 94 publications have accumulated citations equal to or greater than the square of 94). Finally, it was found that within the examined dataset, a total of 1,334 documents (16%) have achieved a minimum of 5 citations (C5), while 691 documents (8%) have attained at least 10 citations (C10), 302 documents (4%) have obtained a minimum of 20 citations (C20), 79 documents (1%) have acquired at least 50 citations (C50), and 31 documents (0.4%) have obtained more than 100 citations (C100).

Table 1 . Descriptive overview of online learning and COVID-19 research (2020–2022).

The most relevant (top-10) highly cited documents in online learning and COVID-19 research in the context of higher education are shown in Table 2 . The overview of the most relevant documents reveals several important topics that were intensively discussed. The first most relevant topic concerns ICT. The COVID-19 pandemic has created significant challenges for higher education, especially for medical and surgical education, which requires personal attendance in clinical activities and labs. Accordingly, several innovative ICT tools (i.e., videoconferencing, social media, and telemedicine) and online learning approaches (i.e., flipped classroom or blended learning and virtual learning) were proposed to address this challenge. It is also stressed that by using appropriately established ICT solutions, online learning can lead to more sustainable education ( Adedoyin and Soykan, 2020 ; Chick et al., 2020 ; Dedeilia et al., 2020 ).

Table 2 . Most relevant documents in online learning and COVID-19 research (2020–2022).

The next top-cited topic relates to pedagogy. The disruption of education around the world due to the COVID-19 pandemic required teachers to possess specific pedagogical content knowledge related to designing and organizing better learning experiences with digital technologies. At the same time, challenges for online assessment and post-pandemic pedagogy are also highlighted ( García Peñalvo et al., 2020 ; Iyer et al., 2020 ; Murphy, 2020 ; Rapanta et al., 2020 ). Finally, life and work is another of the most cited topics. Namely, the COVID-19 pandemic has considerably reshaped education and other aspects of life and work, often also through the perspective of mental health or emotional well-being ( Dwivedi et al., 2020 ; Kapasia et al., 2020 ; Aristovnik et al., 2020a ).

Furthermore, it is noteworthy that all of the highly cited documents were published in 2020. However, it is also evident that there are notable and highly relevant publications that emerged in the second year of the COVID-19 pandemic. Accordingly, there are two documents with a minimum of 100 citations published in 2021. In the COVID-19 pandemic context, Watermeyer et al. (2021) , with 148 citations, examined the implications of digital disruption in universities within the United Kingdom, highlighting the challenges and opportunities arising from the emergency shift to online learning. Meanwhile, Pokhrel and Chhetri (2021) conducted a literature review to assess the impact of the COVID-19 pandemic on teaching and learning.

The scientific production across countries and sources is presented in terms of the number of documents and citations, whereby additional information is provided by a circle’s size, revealing the h-index as a measure of the scientific impact ( Harzing and Van Der Wal, 2009 ) and by its color, presenting the time dimension in scientific production. The most relevant (top-10) highly cited countries in online learning and COVID-19 research are shown in Figure 2 . While the United States of America stands out among all countries, the United Kingdom, China and India have a relatively large number of documents and citations. The findings are similar to those of other bibliometric studies on this topic ( Saqr et al., 2023 ).

Figure 2 . Most relevant countries in online learning and COVID-19 research (2020–2022). Own elaboration based on the Scopus database.

The most relevant (top-10) highly cited sources in online learning and COVID-19 research in the context of higher education are presented in Figure 3 . Despite conference proceedings being prominent in terms of the relatively high number of documents, the most prominent journals, considering the number of citations, are Journal of Chemical Education, with the highest number of citations as well as documents, followed by Sustainability, International Journal of Environmental Research and Public Health, and Education Sciences. More specifically, the most relevant journals address different topics. First, Journal of Chemical Education covers the attempts, successes and failures of distance learning during the COVID-19 pandemic in chemistry education. It covers various topics, including the development of at-home practical activities ( Schultz et al., 2020 ), student engagement and learning ( Perets et al., 2020 ), online assessments ( Nguyen et al., 2020 ) and virtual reality labs ( Williams et al., 2021 ). Further, Sustainability is focused on student and teacher perceptions of e-learning and related challenges ( Khan et al., 2020 ; Aristovnik et al., 2020a ) and sustainability in education during the COVID-19 pandemic ( Sobaih et al., 2020 ) to improve online learning and sustain higher education during uncertain times. Further, the International Journal of Environmental Research and Public Health covers various topics like the health and psychological implications of the COVID-19 pandemic ( Sundarasen et al., 2020 ), including well-being and changes in behavior and habits. Finally, Education Sciences publishes some general research on the challenges and opportunities for online learning ( Almazova et al., 2020 ), including student and teacher experiences ( García-Alberti et al., 2021 ; Müller et al., 2021 ).

Figure 3 . Most relevant sources in online learning and COVID-19 research (2020–2022). Own elaboration based on the Scopus database.

The keyword co-occurrence network is presented in Figure 4 . Note that the nodes indicate keywords and the links the relations of co-occurrence between them. The node size is proportional to the number of keyword occurrences, showing the research intensity (node degree), while the link width is proportional to the co-occurrences between keywords (edge weight). In addition, the node color indicates the cluster to which a particular keyword belongs ( Wang et al., 2020 ; Ravšelj et al., 2022 ). The keyword co-occurrence analysis reveals five research hotspots in online learning in higher education research during the COVID-19 pandemic. These are ICT and pedagogy (red cluster), technology-enhanced education (green cluster), mental health and well-being (blue cluster), student experience (yellow cluster) and curriculum and professional development (purple cluster).

Figure 4 . Keyword co-occurrence network in online learning and COVID-19 research (2020–2022). Own elaboration based on the Scopus database.

A detailed synopsis of the research hotspots, including representative (the most frequent) keywords and documents (with several representative keywords), is presented in Table 3 . The first research hotspot highlights the relevance of ICT and pedagogy in higher education during the COVID-19 pandemic. The most representative documents looked at the quality of online learning mechanisms ( Gritsova and Tissen, 2021 ), active learning activities ( Yan et al., 2021 ) and the role of e-learning departments in controlling the quality of academic processes ( Hamdan et al., 2021 ). The second research hotspot refers to technology-enhanced education from different perspectives, such as opportunities to incorporate technological and curricular innovations ( Shapiro and Reza, 2021 ), the adoption of different virtual experiences such as telehealth and virtual learning ( Kahwash et al., 2021 ), and the utilization of social media to reach higher education students ( Leighton et al., 2021 ). The third research hotspot emphasizes the problem of mental health and well-being issues that became a prevalent topic of discussion during the COVID-19 pandemic. Namely, several studies showed an increase in depression, anxiety and stress levels among higher education students in response to the COVID-19 pandemic ( Abu Kwaik et al., 2021 ; Keskin, 2021 ; Yaghi, 2022 ). The fourth cluster is about student experience during the COVID-19 pandemic with specific focus on the between interaction and online learning satisfaction ( Bawa'aneh, 2021 ; Bismala and Manurung, 2021 ; She et al., 2021 ). The fifth research hotspot underscores the relevance of curriculum and professional development. Several studies described the ways in which courses were adapted to online learning during the COVID-19 pandemic as well as the related challenges and strategies ( Chen et al., 2020 ; Gonzalez and Knecht, 2020 ; Rhile, 2020 ).

Table 3 . Research hotspots based on the author keyword co-occurrence network in online learning and COVID-19 research (2020–2022).

Finally, the three-field plot analysis of the relationship between the main concepts (i.e., ICT tools, online learning approaches, fields of study) is presented in a Sankey diagram shown in Figure 5 . The size of a rectangle corresponds to the number of documents for each theme, while the edge width reflects the inclusion index for connected themes ( Wang et al., 2020 ; Ravšelj et al., 2022 ). These three concepts have been proven to be relevant in the context of online learning. Namely, ICT tools are a precondition for delivering course content through different online learning approaches, while the choice of online learning approaches may depend on the field of study ( Ferri et al., 2020 ). During the COVID-19 pandemic, most attention was devoted to exploring e-learning (a combination of asynchronous and synchronous learning), distance learning (pre-recorded online lectures), followed by virtual learning (real-time online lectures). Since all these online learning approaches limit physical contact between teachers and students, they have been referred to as emergency remote learning approaches ( Hodges et al., 2020 ; Fauzi, 2022 ; Fuchs, 2022 ), while other online learning approaches (computer-based learning, blended learning, m-learning) do not necessarily take place in an online learning environment. The emergency remote learning approaches were primarily supported by several ICT tools, particularly by social media (e.g., Facebook), gamification/simulation and virtual reality (integration of game-like elements into online learning platforms, mobile applications, or virtual reality simulations), Zoom and other videoconferencing platforms, as well as telehealth (for educating health professionals). Regarding the fields of study, e-learning, distance learning and virtual learning were mostly addressed in the context of medical/health education, while computer-based learning (i.e., specific engineering software programs etc.) was examined in the context of engineering education. This implies that the specific requirements of a given field of study often guide the selection of the most suitable online learning approaches ( Fauzi, 2022 ).

Figure 5 . Three-field plot showing the network between ICT tools (left), online approaches (middle), and fields of study (right) (2020–2022). Own elaboration based on the Scopus database.

5. Conclusion

The presented bibliometric study provides several important insights arising from research into online learning during the COVID-19 pandemic. In this period, a large volume of scientific knowledge was produced in the context of education that considered a range of aspects ( Saqr et al., 2023 ). Therefore, a combination of selected bibliometric approaches was utilized to extract some general comprehensive outlines of the global research. The bibliometric analysis revealed the following.

As suggested by the descriptive overview of the state of Educational Research ( Patience et al., 2017 ), the research into online learning during the COVID-19 pandemic is characterized by greater cooperation between authors, which coincides with the general observation that (international) scientific collaboration grew significantly during the pandemic ( Duan and Xia, 2021 ). Further, online learning research during the COVID-19 pandemic is grounded on fewer studies than Educational Research ( Patience et al., 2017 ), which may be explained by the absence of COVID-19-related literature at the time these documents were published. Nevertheless, noting the potential benefits of online learning approaches also when the epidemiological conditions are favorable, this line of research is expected to further develop and be extended in the ensuing years ( Fauzi, 2022 ). The potential benefits refer especially to the development of a blended learning environment, which combines online and traditional learning approaches ( Rasheed et al., 2020 ). The overview of the most relevant documents revealed three topics that were intensively discussed in the academic community, i.e., ICT, pedagogy, and life and work. The COVID-19 pandemic highlighted the importance and role of reliable ICT infrastructure for ensuring effective pedagogy in the online environment, as was needed to prevent the spread of the virus and to protect public health. Apart from the devastating health consequences for those directly affected by the virus and the disrupted educational process, the COVID-19 pandemic also dramatically affected students’ social life and work ( Aristovnik et al., 2020a ). The educational community is increasingly interested in finding ways to respond to crises like the COVID-19 pandemic and develop effective pedagogical practices that assure high-quality and efficient education in the online learning environment ( Zhang et al., 2022 ).

The scientific production of online learning during the COVID-19 pandemic was geographically uneven. The greatest scientific production in terms of citations and number of documents can be observed in the United States, followed by the United Kingdom, China and India. Besides developed English-speaking countries, emerging Asian economies also seem to have played a crucial role in online learning research. Similar findings also emerged from other bibliometric studies on this topic ( Saqr et al., 2023 ). Moreover, despite conference proceedings being prominent in terms of the relatively high number of documents, the most prominent journals, considering the number of citations, are Journal of Chemical Education, Sustainability, International Journal of Environmental Research and Public Health and Education Sciences, indicating that online learning research at the time of the COVID-19 pandemic was primarily published in open-access journals, as already observed in other research ( Zhang et al., 2022 ).

The network analysis revealed five research hotspots in online learning research during the COVID-19 pandemic in the context of higher education: (1) ICT and pedagogy, focused on the quality of online learning mechanisms ( Gritsova and Tissen, 2021 ), active learning activities ( Yan et al., 2021 ) and the role of e-learning departments in controlling the quality of academic processes ( Hamdan et al., 2021 ); technology-enhanced education concentrated on opportunities to incorporate technological and curricular innovations ( Shapiro and Reza, 2021 ), the adoption of different virtual experiences such as telehealth and virtual learning ( Kahwash et al., 2021 ), and the utilization of social media to reach higher education students ( Leighton et al., 2021 ); (2) mental health and well-being issues facing higher education students, including depression, anxiety, and stress levels ( Abu Kwaik et al., 2021 ; Keskin, 2021 ; Yaghi, 2022 ); student experience with specific focus on the between interaction and online learning satisfaction ( Bawa'aneh, 2021 ; Bismala and Manurung, 2021 ; She et al., 2021 ) and (3) curriculum and professional development, focused on the ways in which courses were adapted to online learning during the COVID-19 pandemic as well as the related challenges and strategies ( Chen et al., 2020 ; Gonzalez and Knecht, 2020 ; Rhile, 2020 ).

Further, the COVID-19 pandemic led to the exploration of emergency remote learning approaches such as e-learning, distance learning and virtual learning, which are intended to limit physical contact between teachers and students. These approaches were chiefly supported by several ICT tools, including social media, gamification/simulation, virtual reality, videoconferencing platforms, and telehealth. While computer-based learning, blended learning and m-learning do not necessarily occur in an online learning environment, they may still be suitable for certain fields of study, especially in the post-COVID-19 pandemic period. This implies that the determination of which online learning approach is the most suitable is often guided by the specific requirements of a given field of study ( Fauzi, 2022 ).

Before generalizing these conclusions, it is important to note the limitations of the paper. First, the bibliometric analysis relied on documents indexed in the Scopus database, which might not cover the entire collection of research. Namely, documents that are published in journals indexed in other databases such as Web of Science, Education Research Index, Educational Resources Information Centre, etc. are not included in the analysis. However, to achieve the comparability of bibliometric metrics across documents, the bibliometric metrics are obtained from the single and, in general, broader Scopus database. Given the substantial overlap of documents across different databases of peer-reviewed literature, this limitation might not significantly affect the general observations on global research trends. Nevertheless, to check the robustness of the findings, it is still valuable to consider other bibliometric databases for future research. Second, the bibliometric analysis is conducted the bibliometric is based on a short time period (January 2020 – March 2022), which may also impact the metrics of documents published in closed-access (subscription-based) journals, placing them at a disadvantage compared to documents published in open-access journals. While it is not possible to overcome this limitation at present, conducting a bibliometric study with a longer time span would provide further time-dimensional insights. This would also be beneficial in terms of achieving better comparability between documents published in closed-access and open-access journals. Finally, despite the detailed search queries, some other relevant keywords may have been overlooked in the document search. Finally, the bibliometric method, as a method based on big data analysis, may miss certain highlights from the scientific literature that a systematic literature review would otherwise capture. Therefore it would be beneficial for future bibliometric studies also to incorporate a systematic literature review methodology, as the combined approach can provide a more comprehensive and nuanced understanding of the implications of the COVID-19 pandemic on online learning in higher education.

The bibliometric study provides some possible avenues for future research. First, in future bibliometric studies, it would be beneficial to conduct in-depth analyses of the relevant contexts that have emerged as highly significant in online learning during the pandemic. These include ICT and innovation, mental health and well-being, online learning and engagement, and curriculum and professional development. Examining these contexts more comprehensively can provide valuable insights into the specific dynamics and trends within each area, contributing to a deeper understanding of the implications of online learning during the pandemic. Second, it would be beneficial to conduct separate bibliometric analyses and comparisons to examine the differences between developed and developing countries. This approach can shed light on the unique research trends, contributions, and challenges faced by each group of countries in the context of online learning during the pandemic. This can provide a more nuanced understanding of the global landscape and identify potential areas for collaboration and knowledge sharing between developed and developing countries. Finally, it would be valuable to investigate the long-term impact of rapid publishing in open-access journals on the recognition and dissemination of scholarly findings in the field of online learning in higher education during the pandemic.

From the practical perspective, the COVID-19 pandemic has significantly disrupted higher education, but at the same time, it also accelerated the use of online learning tools in the educational process. Although the COVID-19 pandemic has gradually subsided over time, online learning approaches developed during this period continue to hold relevance and value for future education. Therefore, higher education institutions should prioritize leveraging ICT tools and innovative solutions in their educational delivery, which proved effective during the pandemic. Moreover, higher education institutions should also prioritize adapting appropriate online learning approaches and curricula to align with modern realities and the corresponding fields of study. This adaptation is crucial for enhancing student engagement and ensuring that educational programs remain relevant and responsive to the evolving needs of students in various disciplines.

The findings may help not only the scientific community in detecting research gaps in online learning research during the COVID-19 pandemic but also evidence-based policymaking by assisting in identifying appropriate educational practices in emergency circumstances. Specifically, the findings may help higher education policymakers to address the underlying shortcomings of the existing educational framework exposed by the COVID-19 pandemic and to design proactive mechanisms to deal effectively with such disruptions, thereby enabling them to create a more resilient and adaptable education system that can successfully navigate unforeseen challenges and ensure the continuity of quality higher education in the future.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

Author contributions

AA contributed to the design of the study. DR and LU assisted with the data identification, cleaning, and analysis. DR and KK wrote the manuscript in consultation with AA. All authors contributed to the manuscript’s revision and read and approved the submitted version.

This research and the APC were funded by the Slovenian Research Agency under grant numbers P5-0093 and Z5-4569.

Acknowledgments

The authors acknowledge the financial support from the Slovenian Research Agency (research core funding no. P5-0093 and project no. Z5-4569). A preliminary version of the paper was presented at the International Conference on Information, Communication Technologies in Education (ICICTE) in July 2022. The authors are grateful to colleagues who attended the presentation and provided interesting comments and suggestions. Further, they wish to thank the reviewers for their valuable suggestions and comments.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Abu Kwaik, A., Saleh, R., Danadneh, M., and Kateeb, E. (2021). Stress, anxiety and depression among dental students in times of covid-19 lockdown. Int. J. Dentist. Oral Sci. 8, 1560–1564. doi: 10.19070/2377-8075-21000310

CrossRef Full Text | Google Scholar

Adedoyin, O. B., and Soykan, E. (2020). COVID-19 pandemic and online learning: the challenges and opportunities. Interact. Learn. Environ. 31, 863–875. doi: 10.1080/10494820.2020.1813180

Almazova, N., Krylova, E., Rubtsova, A., and Odinokaya, M. (2020). Challenges and opportunities for Russian higher education amid COVID-19: teachers’ perspective. Educ. Sci. 10:368. doi: 10.3390/educsci10120368

Aristovnik, A., Keržič, D., Ravšelj, D., Tomaževič, N., and Umek, L. (2020a). Impacts of the COVID-19 pandemic on life of higher education students: a global perspective. Sustainability 12:8438. doi: 10.3390/su12208438

Aristovnik, A., Ravšelj, D., and Umek, L. (2020b). A bibliometric analysis of COVID-19 across science and social science research landscape. Sustainability 12:9132. doi: 10.3390/su12219132

Baber, H., Fanea-Ivanovici, M., Lee, Y. T., and Tinmaz, H. (2022). A bibliometric analysis of digital literacy research and emerging themes pre-during COVID-19 pandemic. Inform. Learn. Sci. 123, 214–232. doi: 10.1108/ILS-10-2021-0090

Bawa'aneh, M. S. (2021). Distance learning during COVID-19 pandemic in UAE public schools: student satisfaction, attitudes and challenges. Contemp. Educ. Technol. 13:10872. doi: 10.30935/cedtech/10872

Bilal, H., Hysa, E., Akbar, A., Yasmin, F., Rahman, A. u., and Li, S. (2022). Virtual learning during the COVID-19 pandemic: a bibliometric review and future research agenda. Risk Manag. Healthcare Policy 15, 1353–1368. doi: 10.2147/RMHP.S355895

PubMed Abstract | CrossRef Full Text | Google Scholar

Bismala, L., and Manurung, Y. H. (2021). Student satisfaction in E-learning along the COVID-19 pandemic with importance performance analysis. Int. J. Eval. Res. Educ. 10, 753–759. doi: 10.11591/ijere.v10i3.21467

Bonsaksen, T., Leung, J., Schoultz, M., Thygesen, H., Price, D., Ruffolo, M., et al. (2021). Cross-National Study of worrying, loneliness, and mental health during the COVID-19 pandemic: a comparison between individuals with and without infection in the family. Healthcare 9:903. doi: 10.3390/healthcare9070903

Bozkurt, A. (2022). A retro perspective on blended/hybrid learning: systematic review, mapping and visualization of the scholarly landscape. J. Interact. Media Educ. 2022, 1–15. doi: 10.5334/jime.751

Brika, S. K. M., Chergui, K., Algamdi, A., Musa, A. A., and Zouaghi, R. (2021). E-learning research trends in higher education in light of COVID-19: a bibliometric analysis. Front. Psychol. 12:762819. doi: 10.3389/fpsyg.2021.762819

Campos, E., Daruich, S. D. N., Escamilla, J., and Hosseini, S. (2022). Educational model transition: student evaluation of teaching amid the COVID-19 pandemic. Front. Educ. 7:991654. doi: 10.3389/feduc.2022.991654

Chen, K., Chen, Y., Ling, Y., and Lin, J. (2020). The individual experience of online chemistry teacher education in China: coping with COVID-19 pandemic. J. Chem. Educ. 97, 3265–3270. doi: 10.1021/acs.jchemed.0c00581

Chick, R. C., Clifton, G. T., Peace, K. M., Propper, B. W., Hale, D. F., Alseidi, A. A., et al. (2020). Using technology to maintain the education of residents during the COVID-19 pandemic. J. Surg. Educ. 77, 729–732. doi: 10.1016/j.jsurg.2020.03.018

Dedeilia, A., Papapanou, M., Papadopoulos, A. N., Karela, N. R., Androutsou, A., Mitsopoulou, D., et al. (2023). Health worker education during the COVID-19 pandemic: global disruption, responses and lessons for the future—a systematic review and meta-analysis. Hum. Resour. Health 21, 13–35. doi: 10.1186/s12960-023-00799-4

Dedeilia, A., Sotiropoulos, M. G., Hanrahan, J. G., Janga, D., Dedeilias, P., and Sideris, M. (2020). Medical and surgical education challenges and innovations in the COVID-19 era: a systematic review. In Vivo 34, 1603–1611. doi: 10.21873/invivo.11950

Djeki, E., Dégila, J., Bondiombouy, C., and Alhassan, M. H. (2022). E-learning bibliometric analysis from 2015 to 2020. J. Comput. Educ. 9, 727–754. doi: 10.1007/s40692-021-00218-4

Duan, D., and Xia, Q. (2021). Evolution of scientific collaboration on COVID-19: a bibliometric analysis. Learn. Publish. 34, 429–441. doi: 10.1002/leap.1382

Dwivedi, Y. K., Hughes, D. L., Coombs, C., Constantiou, I., Duan, Y., Edwards, J. S., et al. (2020). Impact of COVID-19 pandemic on information management research and practice: transforming education, work and life. Int. J. Inf. Manag. 55:102211. doi: 10.1016/j.ijinfomgt.2020.102211

Elmer, T., Mepham, K., and Stadtfeld, C. (2020). Students under lockdown: comparisons of students' social networks and mental health before and during the COVID-19 crisis in Switzerland. PLoS One 15:e0236337. doi: 10.1371/journal.pone.0236337

Falagas, M. E., Pitsouni, E. I., Malietzis, G. A., and Pappas, G. (2008). Comparison of PubMed, Scopus, web of science, and Google scholar: strengths and weaknesses. FASEB J. 22, 338–342. doi: 10.1096/fj.07-9492LSF

Fauzi, M. A. (2022). E-learning in higher education institutions during COVID-19 pandemic: current and future trends through bibliometric analysis. Heliyon 8:e09433. doi: 10.1016/j.heliyon.2022.e09433

Ferri, F., Grifoni, P., and Guzzo, T. (2020). Online learning and emergency remote teaching: opportunities and challenges in emergency situations. Societies 10:86. doi: 10.3390/soc10040086

Fried, E. I., Papanikolaou, F., and Epskamp, S. (2021). Mental health and social contact during the COVID-19 pandemic: an ecological momentary assessment study. Clin. Psychol. Sci. 10, 340–354. doi: 10.1177/21677026211017839

Fuchs, K. (2022). The difference between emergency remote teaching and e-learning. Front. Educ. 7:921332. doi: 10.3389/feduc.2022.921332

Gao, Y., Wong, S. L., Khambari, M. N., and Noordin, N. (2022). A bibliometric analysis of the scientific production of e-learning in higher education (1998-2020). Int. J. Inform. Educ. Technol. 12, 390–399. doi: 10.18178/ijiet.2022.12.5.1632

García Peñalvo, F. J., Corell Almuzara, A., Abella García, V., and Grande de Prado, M. (2020). La evaluación online en la educación superior en tempos’ de la COVID-19. Educ. Knowl. Soc. 21:23086. doi: 10.14201/eks.23086

García-Alberti, M., Suárez, F., Chiyón, I., and Mosquera Feijoo, J. C. (2021). Challenges and experiences of online evaluation in courses of civil engineering during the lockdown learning due to the COVID-19 pandemic. Educ. Sci. 11:59. doi: 10.3390/educsci11020059

Gavriluţă, C., Dalban, C. M., and Ioan, B. G. (2022). Educational, emotional, and social impact of the emergency state of COVID-19 on Romanian university students. Int. J. Environ. Res. Public Health 19:3990. doi: 10.3390/ijerph19073990

Gonzalez, C., and Knecht, L. D. (2020). Strategies employed in transitioning multi-instructor, multisection introductory general and organic chemistry courses from face-to-face to online learning. J. Chem. Educ. 97, 2871–2877. doi: 10.1021/acs.jchemed.0c00670

Gritsova, O. A., and Tissen, E. V. (2021). Quality assessment of online learning in regional higher education systems. Econ. Regions 17, 929–943. doi: 10.17059/ekon.reg.2021-3-15

Gurcan, F., Dalveren, G. G. M., and Derawi, M. (2022). COVID-19 and E-learning: an exploratory analysis of research topics and interests in E-learning during the pandemic. IEEE Access 10, 123349–123357. doi: 10.1109/ACCESS.2022.3224034

Hamdan, R., Ashour, W., and Daher, W. (2021). The role of the e-learning departments in controlling the quality of electronic assessments in Palestinian universities during the COVID-19 pandemic. Sustainability 13:12021. doi: 10.3390/su132112021

Harzing, A. W., and Van Der Wal, R. (2009). A Google scholar h-index for journals: an alternative metric to measure journal impact in economics and business. J. Am. Soc. Inf. Sci. Technol. 60, 41–46. doi: 10.1002/asi.20953

Hodges, C. B., Moore, S., Lockee, B. B., Trust, T., and Bond, M. A. (2020). The difference between emergency remote teaching and online learning. Virginia Tech Online Available online at: http://hdl.handle.net/10919/104648 (Accessed February 15, 2023).

Google Scholar

Hunter, J. D. (2007). Matplotlib: a 2D graphics environment. Comput. Sci. Eng. 9, 90–95. doi: 10.1109/MCSE.2007.55

Iyer, P., Aziz, K., and Ojcius, D. M. (2020). Impact of COVID-19 on dental education in the United States. J. Dent. Educ. 84, 718–722. doi: 10.1002/jdd.12163

James, N., and Thériault, V. (2020). Adult education in times of the COVID-19 pandemic: inequalities, changes, and resilience. Stud. Educ. Adults 52, 129–133. doi: 10.1080/02660830.2020.1811474

Kahwash, B. M., Deshpande, D. R., Guo, C., Panganiban, C. M., Wangberg, H., and Craig, T. J. (2021). Allergy/immunology trainee experiences during the COVID-19 pandemic: AAAAI work group report of the fellows-in-training committee. J. Allergy Clin. Immunol. Pract. 9, 1–6.e1. doi: 10.1016/j.jaip.2020.09.036

Kapasia, N., Paul, P., Roy, A., Saha, J., Zaveri, A., Mallick, R., et al. (2020). Impact of lockdown on learning status of undergraduate and postgraduate students during COVID-19 pandemic in West Bengal, India. Child Youth Serv. Rev. 116:105194. doi: 10.1016/j.childyouth.2020.105194

Kartimi, K., Yunita, Y., Addiin, I., and Shidiq, A. S. (2022). A bibliometric analysis on chemistry virtual laboratory. Educación Química 33:194. doi: 10.22201/fq.18708404e.2022.2.80579

Keržič, D., Alex, J. K., Pamela Balbontín Alvarado, R., da Silva Bezerra, D., Cheraghi, M., Dobrowolska, B., et al. (2021). Academic student satisfaction and perceived performance in the e-learning environment during the COVID-19 pandemic: evidence across ten countries. PLoS One 16:e0258807. doi: 10.1371/journal.pone.0258807

Keržič, D., Umek, L., Tomaževič, N., and Aristovnik, A. (2022). E-learning acceptance of public administration students in Slovenia in comparison with Europe: differences between the first and second waves of the COVID-19 pandemic. EGPA 2022 Conference, 6–9 September 2022, Lisbon, Portugal.

Keskin, G. (2021). Self-report measurement of depression, anxiety, and stress caused by covid-19 pandemic in senior undergraduate dental students. Pesquisa Brasileira em Odontopediatria e Clínica Integrada 21:102. doi: 10.1590/pboci.2021.102

Khan, M. A., Nabi, M. K., Khojah, M., and Tahir, M. (2020). Students’ perception towards e-learning during COVID-19 pandemic in India: an empirical study. Sustainability 13:57. doi: 10.3390/su13010057

Küçük-Avci, Ş., Topal, M., and İstanbullu, A. (2022). The effects of the COVID-19 pandemic on distance education in higher education: a bibliometric analysis study. Croatian J. Educ. 24, 457–488. doi: 10.15516/cje.v24i2.4534

Leighton, K., Kardong-Edgren, S., Schneidereith, T., and Foisy-Doll, C. (2021). Using social media and snowball sampling as an alternative recruitment strategy for research. Clin. Simul. Nurs. 55, 37–42. doi: 10.1016/j.ecns.2021.03.006

McKinney, W. (2012). Python for data analysis: Data wrangling with pandas, NumPy, and IPython . Sebastopol, CA: O'Reilly Media, Inc.

Misiejuk, K., Ness, I. J., Gray, R., and Wasson, B. (2023). Changes in online course designs: before, during, and after the pandemic. Front. Educ. 7:996006. doi: 10.3389/feduc.2022.996006

Mongeon, P., and Paul-Hus, A. (2016). The journal coverage of web of science and Scopus: a comparative analysis. Scientometrics 106, 213–228. doi: 10.1007/s11192-015-1765-5

Müller, A. M., Goh, C., Lim, L. Z., and Gao, X. (2021). COVID-19 emergency elearning and beyond: experiences and perspectives of university educators. Educ. Sci. 11:19. doi: 10.3390/educsci11010019

Murphy, M. P. (2020). COVID-19 and emergency eLearning: consequences of the securitization of higher education for post-pandemic pedagogy. Contemp. Security Policy 41, 492–505. doi: 10.1080/13523260.2020.1761749

Navarro-Espinosa, J. A., Vaquero-Abellán, M., Perea-Moreno, A.-J., Pedrós-Pérez, G., Aparicio-Martínez, P., and Martínez-Jiménez, M. P. (2021). The influence of technology on mental well-being of STEM teachers at university level: COVID-19 as a stressor. Int. J. Environ. Res. Public Health 18:9605. doi: 10.3390/ijerph18189605

Nguyen, J. G., Keuseman, K. J., and Humston, J. J. (2020). Minimize online cheating for online assessments during COVID-19 pandemic. J. Chem. Educ. 97, 3429–3435. doi: 10.1021/acs.jchemed.0c00790

Ochnik, D., Rogowska, A. M., Kuśnierz, C., Jakubiak, M., Wierzbik-Strońska, M., Schütz, A., et al. (2021). Exposure to COVID-19 during the first and the second wave of the pandemic and coronavirus-related PTSD risk among university students from six countries-a repeated cross-sectional study. J. Clin. Med. 10:5564. doi: 10.3390/jcm10235564

Okoro, C., Owojori, O. M., and Umeokafor, N. (2022). The developmental trajectory of a decade of research on mental health and well-being amongst graduate students: a bibliometric analysis. Int. J. Environ. Res. Public Health 19:4929. doi: 10.3390/ijerph19094929

Pandey, K., and Panchal, R. (2020). A study of real world data visualization of COVID-19 dataset using Python. Int. J. Manag. Human. 4, 104–107. doi: 10.35940/ijmh.H0834.044820

Patience, G. S., Patience, C. A., Blais, B., and Bertrand, F. (2017). Citation analysis of scientific categories. Heliyon 3:e00300. doi: 10.1016/j.heliyon.2017.e00300

Perets, E. A., Chabeda, D., Gong, A. Z., Huang, X., Fung, T. S., Ng, K. Y., et al. (2020). Impact of the emergency transition to remote teaching on student engagement in a non-STEM undergraduate chemistry course in the time of COVID-19. J. Chem. Educ. 97, 2439–2447. doi: 10.1021/acs.jchemed.0c00879

Pham, P. T., Lien, D. T. H., Kien, H. C., Chi, N. H., Tinh, P. T., Do, T., et al. (2022). Learning management system in developing countries: a bibliometric analysis between 2005 and 2020. Europ. J. Educ. Res. 11, 1363–1377. doi: 10.12973/eu-jer.11.3.1363

Pokhrel, S., and Chhetri, R. (2021). A literature review on impact of COVID-19 pandemic on teaching and learning. High. Educ. Future 8, 133–141. doi: 10.1177/234763112098348

Raccanello, D., Balbontín-Alvarado, R., da Silva Bezerra, D., Burro, R., Cheraghi, M., Dobrowolska, B., et al. (2022). Higher education students’ achievement emotions and their antecedents in e-learning amid COVID-19 pandemic: a multi-country survey. Learn. Instr. 80:101629. doi: 10.1016/j.learninstruc.2022.101629

Rapanta, C., Botturi, L., Goodyear, P., Guàrdia, L., and Koole, M. (2020). Online university teaching during and after the COVID-19 crisis: refocusing teacher presence and learning activity. Postdigital Sci. Educ. 2, 923–945. doi: 10.1007/s42438-020-00155-y

Rasheed, R. A., Kamsin, A., and Abdullah, N. A. (2020). Challenges in the online component of blended learning: a systematic review. Comput. Educ. 144:103701. doi: 10.1016/j.compedu.2019.103701

Rasli, A., Tee, M., Ling, L. Y., Cheak, T. Z., and Hui, S. E. (2022). Post COVID-19 strategies for higher education institutions in dealing with unknown and uncertainties. Front. Educ. 7:992063. doi: 10.3389/feduc.2022.992063

Ravšelj, D., Umek, L., Todorovski, L., and Aristovnik, A. (2022). A review of digital era governance research in the first two decades: a bibliometric study. Future Internet 14:126. doi: 10.3390/fi14050126

Rhile, I. J. (2020). Course redesign for college general chemistry during the COVID-19 pandemic. J. Chem. Educ. 97, 2857–2862. doi: 10.1021/acs.jchemed.0c00618

Rogers, G., Szomszor, M., and Adams, J. (2020). Sample size in bibliometric analysis. Scientometrics 125, 777–794. doi: 10.1007/s11192-020-03647-7

Rojas-Sánchez, M. A., Palos-Sánchez, P. R., and Folgado-Fernández, J. A. (2022). Systematic literature review and bibliometric analysis on virtual reality and education. Educ. Inf. Technol. 28, 155–192. doi: 10.1007/s10639-022-11167-5

Saqr, M., Raspopovic Milic, M., Pancheva, K., Jovic, J., Peltekova, E. V., and Conde, M. Á. (2023). A multimethod synthesis of COVID-19 education research: the tightrope between covidization and meaningfulness. Univ. Access Inf. Soc. 1-14, 1–14. doi: 10.1007/s10209-023-00989-w

Schultz, M., Callahan, D. L., and Miltiadous, A. (2020). Development and use of kitchen chemistry home practical activities during unanticipated campus closures. J. Chem. Educ. 97, 2678–2684. doi: 10.1021/acs.jchemed.0c00620

Shapiro, H., and Reza, N. (2021). Cardiovascular medical education during the coronavirus disease 2019 pandemic: challenges, adaptations, and considerations for the future. US Cardiol. 15:25. doi: 10.15420/usc.2020.25

She, L., Ma, L., Jan, A., Sharif Nia, H., and Rahmatpour, P. (2021). Online learning satisfaction during COVID-19 pandemic among Chinese university students: the serial mediation model. Front. Psychol. 12:743936. doi: 10.3389/fpsyg.2021.743936

Sobaih, A. E. E., Hasanein, A. M., and Abu Elnasr, A. E. (2020). Responses to COVID-19 in higher education: social media usage for sustaining formal academic communication in developing countries. Sustainability 12:6520. doi: 10.3390/su12166520x

Sundarasen, S., Chinna, K., Kamaludin, K., Nurunnabi, M., Baloch, G. M., Khoshaim, H. B., et al. (2020). Psychological impact of COVID-19 and lockdown among university students in Malaysia: implications and policy recommendations. Int. J. Environ. Res. Public Health 17:6206. doi: 10.3390/ijerph17176206

Tang, K. H. D. (2023). Impacts of COVID-19 on primary, secondary and tertiary education: a comprehensive review and recommendations for educational practices. Educ. Res. Policy Prac. 22, 23–61. doi: 10.1007/s10671-022-09319-y

Tlili, A., Altınay, F., Altınay, Z., Aydın, C. H., Huang, R., and Sharma, R. (2022). Reflections on massive open online courses (Moocs) during the COVID-19 pandemic: a bibliometric mapping analysis. Turk. Online J. Dist. Educ. 23, 1–17. doi: 10.17718/tojde.1137107

Vaicondam, Y., Sikandar, H., Irum, S., Khan, N., and Qureshi, M. I. (2022). Research landscape of digital learning over the past 20 years: a bibliometric and visualisation analysis. Int. J. Online Biomed. Eng. 18, 4–22. doi: 10.3991/ijoe.v18i08.31963

Van der Graaf, L., Dunajeva, J., Siarova, H., and Bankauskaite, R. (2021). Research for CULT committee – Education and youth in post-COVID-19 Europe – Crisis effects and policy recommendations. European Parliament, Policy Department for Structural and Cohesion Policies, Brussels. Available at: http://www.europarl.europa.eu/thinktank/en/document.html?reference=IPOL_STU(2021)690872 (Accessed February 15, 2023).

Van Eck, N., and Waltman, L. (2010). Software survey: VOSviewer, a computer program for bibliometric mapping. Scientometrics 84, 523–538. doi: 10.1007/s11192-009-0146-3

Vera-Baceta, M. A., Thelwall, M., and Kousha, K. (2019). Web of science and Scopus language coverage. Scientometrics 121, 1803–1813. doi: 10.1007/s11192-019-03264-z

Waller, R., Hodge, S., Holford, J., Milana, M., and Webb, S. (2020). Lifelong education, social inequality and the COVID-19 health pandemic. Int. J. Lifelong Educ. 39, 243–246. doi: 10.1080/02601370.2020.1790267

Wang, C., Lim, M. K., Zhao, L., Tseng, M. L., Chien, C. F., and Lev, B. (2020). The evolution of omega-the international journal of management science over the past 40 years: a bibliometric overview. Omega 93:102098. doi: 10.1016/j.omega.2019.08.005

Watermeyer, R., Crick, T., Knight, C., and Goodall, J. (2021). COVID-19 and digital disruption in UK universities: afflictions and affordances of emergency online migration. High. Educ. 81, 623–641. doi: 10.1007/s10734-020-00561-y

Williams, N. D., Gallardo-Williams, M. T., Griffith, E. H., and Bretz, S. L. (2021). Investigating meaningful learning in virtual reality organic chemistry laboratories. J. Chem. Educ. 99, 1100–1105. doi: 10.1021/acs.jchemed.1c00476

Yaghi, A. (2022). Impact of online education on anxiety and stress among undergraduate public affairs students: a longitudinal study during the COVID-19 pandemic. J. Public Affairs Educ. 28, 91–108. doi: 10.1080/15236803.2021.1954469

Yan, Y., Cheng, X., Zhou, C., Yang, X., and Li, Y. Q. (2021). The perceptions of anatomy teachers for different majors during the COVID-19 pandemic: a national Chinese survey. Med. Educ. Online 26:1897267. doi: 10.1080/10872981.2021.1897267

Yan, H., Rahgozar, A., Sethuram, C., Karunananthan, S., Archibald, D., Bradley, L., et al. (2022). Natural language processing to identify digital learning tools in postgraduate family medicine: protocol for a scoping review. JMIR Res. Protocols 11:e34575. doi: 10.2196/34575

Zhang, L., Carter, R. A. Jr., Qian, X., Yang, S., Rujimora, J., and Wen, S. (2022). Academia’s responses to crisis: a bibliometric analysis of literature on online learning in higher education during COVID-19. Br. J. Educ. Technol. 53, 620–646. doi: 10.1111/bjet.13191

Keywords: online learning, e-learning, higher education, bibliometrics, mapping, visualization, VOSviewer, COVID-19

Citation: Aristovnik A, Karampelas K, Umek L and Ravšelj D (2023) Impact of the COVID-19 pandemic on online learning in higher education: a bibliometric analysis. Front. Educ . 8:1225834. doi: 10.3389/feduc.2023.1225834

Received: 19 May 2023; Accepted: 14 July 2023; Published: 03 August 2023.

Reviewed by:

Copyright © 2023 Aristovnik, Karampelas, Umek and Ravšelj. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Aleksander Aristovnik, [email protected] ; Dejan Ravšelj, [email protected]

This article is part of the Research Topic

Increased Quality Education Through Cross-Campus Learning Environments

• Help & FAQ

Students’ experience of online learning during the COVID-19 pandemic: a province-wide survey study

• Department of Data Science & AI
• Portfolio of the Deputy Vice-Chancellor and Vice-President (Education)

Research output : Contribution to journal › Article › Research › peer-review

Online learning is currently adopted by educational institutions worldwide to provide students with ongoing education during the COVID-19 pandemic. Even though online learning research has been advancing in uncovering student experiences in various settings (i.e., tertiary, adult, and professional education), very little progress has been achieved in understanding the experience of the K-12 student population, especially when narrowed down to different school-year segments (i.e., primary and secondary school students). This study explores how students at different stages of their K-12 education reacted to the mandatory full-time online learning during the COVID-19 pandemic. For this purpose, we conducted a province-wide survey study in which the online learning experience of 1,170,769 Chinese students was collected from the Guangdong Province of China. We performed cross-tabulation and Chi-square analysis to compare students’ online learning conditions, experiences, and expectations. Results from this survey study provide evidence that students’ online learning experiences are significantly different across school years. Foremost, policy implications were made to advise government authorises and schools on improving the delivery of online learning, and potential directions were identified for future research into K-12 online learning. Practitioner notes What is already known about this topic Online learning has been widely adopted during the COVID-19 pandemic to ensure the continuation of K-12 education. Student success in K-12 online education is substantially lower than in conventional schools. Students experienced various difficulties related to the delivery of online learning. What this paper adds Provide empirical evidence for the online learning experience of students in different school years. Identify the different needs of students in primary, middle, and high school. Identify the challenges of delivering online learning to students of different age. Implications for practice and/or policy Authority and schools need to provide sufficient technical support to students in online learning. The delivery of online learning needs to be customised for students in different school years.

• learner attitudes/perceptions
• primary education
• questionnaire
• secondary education

Access to Document

• 10.1111/bjet.13102
• 340340981-oa Final published version, 1.6 MB

Other files and links

• Link to publication in Scopus

T1 - Students’ experience of online learning during the COVID-19 pandemic

T2 - a province-wide survey study

AU - Yan, Lixiang

AU - Whitelock-Wainwright, Alexander

AU - Guan, Quanlong

AU - Wen, Gangxin

AU - Gašević, Dragan

AU - Chen, Guanliang

N1 - Funding Information: This work is supported by the National Natural Science Foundation of China (62077028, 61877029), the Science and Technology Planning Project of Guangdong (2020B0909030005, 2020B1212030003, 2020ZDZX3013, 2019B1515120010, 2018KTSCX016, 2019A050510024), the Science and Technology Planning Project of Guangzhou (201902010041), and the Fundamental Research Funds for the Central Universities (21617408, 21619404). Publisher Copyright: © 2021 British Educational Research Association Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

PY - 2021/9

Y1 - 2021/9

N2 - Online learning is currently adopted by educational institutions worldwide to provide students with ongoing education during the COVID-19 pandemic. Even though online learning research has been advancing in uncovering student experiences in various settings (i.e., tertiary, adult, and professional education), very little progress has been achieved in understanding the experience of the K-12 student population, especially when narrowed down to different school-year segments (i.e., primary and secondary school students). This study explores how students at different stages of their K-12 education reacted to the mandatory full-time online learning during the COVID-19 pandemic. For this purpose, we conducted a province-wide survey study in which the online learning experience of 1,170,769 Chinese students was collected from the Guangdong Province of China. We performed cross-tabulation and Chi-square analysis to compare students’ online learning conditions, experiences, and expectations. Results from this survey study provide evidence that students’ online learning experiences are significantly different across school years. Foremost, policy implications were made to advise government authorises and schools on improving the delivery of online learning, and potential directions were identified for future research into K-12 online learning. Practitioner notes What is already known about this topic Online learning has been widely adopted during the COVID-19 pandemic to ensure the continuation of K-12 education. Student success in K-12 online education is substantially lower than in conventional schools. Students experienced various difficulties related to the delivery of online learning. What this paper adds Provide empirical evidence for the online learning experience of students in different school years. Identify the different needs of students in primary, middle, and high school. Identify the challenges of delivering online learning to students of different age. Implications for practice and/or policy Authority and schools need to provide sufficient technical support to students in online learning. The delivery of online learning needs to be customised for students in different school years.

AB - Online learning is currently adopted by educational institutions worldwide to provide students with ongoing education during the COVID-19 pandemic. Even though online learning research has been advancing in uncovering student experiences in various settings (i.e., tertiary, adult, and professional education), very little progress has been achieved in understanding the experience of the K-12 student population, especially when narrowed down to different school-year segments (i.e., primary and secondary school students). This study explores how students at different stages of their K-12 education reacted to the mandatory full-time online learning during the COVID-19 pandemic. For this purpose, we conducted a province-wide survey study in which the online learning experience of 1,170,769 Chinese students was collected from the Guangdong Province of China. We performed cross-tabulation and Chi-square analysis to compare students’ online learning conditions, experiences, and expectations. Results from this survey study provide evidence that students’ online learning experiences are significantly different across school years. Foremost, policy implications were made to advise government authorises and schools on improving the delivery of online learning, and potential directions were identified for future research into K-12 online learning. Practitioner notes What is already known about this topic Online learning has been widely adopted during the COVID-19 pandemic to ensure the continuation of K-12 education. Student success in K-12 online education is substantially lower than in conventional schools. Students experienced various difficulties related to the delivery of online learning. What this paper adds Provide empirical evidence for the online learning experience of students in different school years. Identify the different needs of students in primary, middle, and high school. Identify the challenges of delivering online learning to students of different age. Implications for practice and/or policy Authority and schools need to provide sufficient technical support to students in online learning. The delivery of online learning needs to be customised for students in different school years.

KW - e-learning

KW - learner attitudes/perceptions

KW - primary education

KW - questionnaire

KW - secondary education

UR - http://www.scopus.com/inward/record.url?scp=85105064273&partnerID=8YFLogxK

U2 - 10.1111/bjet.13102

DO - 10.1111/bjet.13102

M3 - Article

C2 - 34219755

AN - SCOPUS:85105064273

SN - 0007-1013

JO - British Journal of Educational Technology

JF - British Journal of Educational Technology

The pandemic has had devastating impacts on learning. What will it take to help students catch up?

Subscribe to the brown center on education policy newsletter, megan kuhfeld , megan kuhfeld senior research scientist - nwea @megankuhfeld jim soland , jim soland assistant professor, school of education and human development - university of virginia, affiliated research fellow - nwea @jsoland karyn lewis , and karyn lewis director, center for school and student progress - nwea @karynlew emily morton emily morton research scientist - nwea @emily_r_morton.

March 3, 2022

As we reach the two-year mark of the initial wave of pandemic-induced school shutdowns, academic normalcy remains out of reach for many students, educators, and parents. In addition to surging COVID-19 cases at the end of 2021, schools have faced severe staff shortages , high rates of absenteeism and quarantines , and rolling school closures . Furthermore, students and educators continue to struggle with mental health challenges , higher rates of violence and misbehavior , and concerns about lost instructional time .

As we outline in our new research study released in January, the cumulative impact of the COVID-19 pandemic on students’ academic achievement has been large. We tracked changes in math and reading test scores across the first two years of the pandemic using data from 5.4 million U.S. students in grades 3-8. We focused on test scores from immediately before the pandemic (fall 2019), following the initial onset (fall 2020), and more than one year into pandemic disruptions (fall 2021).

Average fall 2021 math test scores in grades 3-8 were 0.20-0.27 standard deviations (SDs) lower relative to same-grade peers in fall 2019, while reading test scores were 0.09-0.18 SDs lower. This is a sizable drop. For context, the math drops are significantly larger than estimated impacts from other large-scale school disruptions, such as after Hurricane Katrina—math scores dropped 0.17 SDs in one year for New Orleans evacuees .

Even more concerning, test-score gaps between students in low-poverty and high-poverty elementary schools grew by approximately 20% in math (corresponding to 0.20 SDs) and 15% in reading (0.13 SDs), primarily during the 2020-21 school year. Further, achievement tended to drop more between fall 2020 and 2021 than between fall 2019 and 2020 (both overall and differentially by school poverty), indicating that disruptions to learning have continued to negatively impact students well past the initial hits following the spring 2020 school closures.

These numbers are alarming and potentially demoralizing, especially given the heroic efforts of students to learn and educators to teach in incredibly trying times. From our perspective, these test-score drops in no way indicate that these students represent a “ lost generation ” or that we should give up hope. Most of us have never lived through a pandemic, and there is so much we don’t know about students’ capacity for resiliency in these circumstances and what a timeline for recovery will look like. Nor are we suggesting that teachers are somehow at fault given the achievement drops that occurred between 2020 and 2021; rather, educators had difficult jobs before the pandemic, and now are contending with huge new challenges, many outside their control.

Clearly, however, there’s work to do. School districts and states are currently making important decisions about which interventions and strategies to implement to mitigate the learning declines during the last two years. Elementary and Secondary School Emergency Relief (ESSER) investments from the American Rescue Plan provided nearly $200 billion to public schools to spend on COVID-19-related needs. Of that sum, $22 billion is dedicated specifically to addressing learning loss using “evidence-based interventions” focused on the “ disproportionate impact of COVID-19 on underrepresented student subgroups. ” Reviews of district and state spending plans (see Future Ed , EduRecoveryHub , and RAND’s American School District Panel for more details) indicate that districts are spending their ESSER dollars designated for academic recovery on a wide variety of strategies, with summer learning, tutoring, after-school programs, and extended school-day and school-year initiatives rising to the top.

Comparing the negative impacts from learning disruptions to the positive impacts from interventions

To help contextualize the magnitude of the impacts of COVID-19, we situate test-score drops during the pandemic relative to the test-score gains associated with common interventions being employed by districts as part of pandemic recovery efforts. If we assume that such interventions will continue to be as successful in a COVID-19 school environment, can we expect that these strategies will be effective enough to help students catch up? To answer this question, we draw from recent reviews of research on high-dosage tutoring , summer learning programs , reductions in class size , and extending the school day (specifically for literacy instruction) . We report effect sizes for each intervention specific to a grade span and subject wherever possible (e.g., tutoring has been found to have larger effects in elementary math than in reading).

Figure 1 shows the standardized drops in math test scores between students testing in fall 2019 and fall 2021 (separately by elementary and middle school grades) relative to the average effect size of various educational interventions. The average effect size for math tutoring matches or exceeds the average COVID-19 score drop in math. Research on tutoring indicates that it often works best in younger grades, and when provided by a teacher rather than, say, a parent. Further, some of the tutoring programs that produce the biggest effects can be quite intensive (and likely expensive), including having full-time tutors supporting all students (not just those needing remediation) in one-on-one settings during the school day. Meanwhile, the average effect of reducing class size is negative but not significant, with high variability in the impact across different studies. Summer programs in math have been found to be effective (average effect size of .10 SDs), though these programs in isolation likely would not eliminate the COVID-19 test-score drops.

Figure 1: Math COVID-19 test-score drops compared to the effect sizes of various educational interventions

Source: COVID-19 score drops are pulled from Kuhfeld et al. (2022) Table 5; reduction-in-class-size results are from pg. 10 of Figles et al. (2018) Table 2; summer program results are pulled from Lynch et al (2021) Table 2; and tutoring estimates are pulled from Nictow et al (2020) Table 3B. Ninety-five percent confidence intervals are shown with vertical lines on each bar.

Notes: Kuhfeld et al. and Nictow et al. reported effect sizes separately by grade span; Figles et al. and Lynch et al. report an overall effect size across elementary and middle grades. We were unable to find a rigorous study that reported effect sizes for extending the school day/year on math performance. Nictow et al. and Kraft & Falken (2021) also note large variations in tutoring effects depending on the type of tutor, with larger effects for teacher and paraprofessional tutoring programs than for nonprofessional and parent tutoring. Class-size reductions included in the Figles meta-analysis ranged from a minimum of one to minimum of eight students per class.

Figure 2 displays a similar comparison using effect sizes from reading interventions. The average effect of tutoring programs on reading achievement is larger than the effects found for the other interventions, though summer reading programs and class size reduction both produced average effect sizes in the ballpark of the COVID-19 reading score drops.

Figure 2: Reading COVID-19 test-score drops compared to the effect sizes of various educational interventions

Source: COVID-19 score drops are pulled from Kuhfeld et al. (2022) Table 5; extended-school-day results are from Figlio et al. (2018) Table 2; reduction-in-class-size results are from pg. 10 of Figles et al. (2018) ; summer program results are pulled from Kim & Quinn (2013) Table 3; and tutoring estimates are pulled from Nictow et al (2020) Table 3B. Ninety-five percent confidence intervals are shown with vertical lines on each bar.

Notes: While Kuhfeld et al. and Nictow et al. reported effect sizes separately by grade span, Figlio et al. and Kim & Quinn report an overall effect size across elementary and middle grades. Class-size reductions included in the Figles meta-analysis ranged from a minimum of one to minimum of eight students per class.

There are some limitations of drawing on research conducted prior to the pandemic to understand our ability to address the COVID-19 test-score drops. First, these studies were conducted under conditions that are very different from what schools currently face, and it is an open question whether the effectiveness of these interventions during the pandemic will be as consistent as they were before the pandemic. Second, we have little evidence and guidance about the efficacy of these interventions at the unprecedented scale that they are now being considered. For example, many school districts are expanding summer learning programs, but school districts have struggled to find staff interested in teaching summer school to meet the increased demand. Finally, given the widening test-score gaps between low- and high-poverty schools, it’s uncertain whether these interventions can actually combat the range of new challenges educators are facing in order to narrow these gaps. That is, students could catch up overall, yet the pandemic might still have lasting, negative effects on educational equality in this country.

Given that the current initiatives are unlikely to be implemented consistently across (and sometimes within) districts, timely feedback on the effects of initiatives and any needed adjustments will be crucial to districts’ success. The Road to COVID Recovery project and the National Student Support Accelerator are two such large-scale evaluation studies that aim to produce this type of evidence while providing resources for districts to track and evaluate their own programming. Additionally, a growing number of resources have been produced with recommendations on how to best implement recovery programs, including scaling up tutoring , summer learning programs , and expanded learning time .

Ultimately, there is much work to be done, and the challenges for students, educators, and parents are considerable. But this may be a moment when decades of educational reform, intervention, and research pay off. Relying on what we have learned could show the way forward.

Related Content

Megan Kuhfeld, Jim Soland, Beth Tarasawa, Angela Johnson, Erik Ruzek, Karyn Lewis

December 3, 2020

Lindsay Dworkin, Karyn Lewis

October 13, 2021

Education Policy K-12 Education

Governance Studies

Brown Center on Education Policy

Katharine Meyer

May 7, 2024

Jamie Klinenberg, Jon Valant, Nicolas Zerbino

Thinley Choden

May 3, 2024

• Share full article

Advertisement

Supported by

David French

Colleges Have Gone off the Deep End. There Is a Way Out.

By David French

Opinion Columnist

I had my head in a law book when I heard the drums. That was the sound of the first campus protest I ever experienced. I’d come to Harvard Law School in the fall of 1991 as a graduate of a small, very conservative Christian college in Nashville. Many of my college classmates had passionate religious and political commitments, but street protest was utterly alien to the Christian culture of the school. We were rule followers, and public protest looked a bit too much like anarchy for our tastes.

But Harvard was different. The law school was every bit as progressive as my college was conservative, and protest was part of the fabric of student life, especially then. This is the era when a writer for GQ magazine, John Sedgwick, called the law school “ Beirut on the Charles ” because it was torn apart by disputes over race and sex. There were days when campus protests were festive, almost celebratory. There were other days when the campus was seething with rage and fury.

That first protest was in support of faculty diversity, and it was relatively benign. I walked outside and followed the sound of the drums. A group of roughly 100 protesters was marching in front of the law school library, and soon they were joined by an allied group of similar size from Harvard’s John F. Kennedy School of Government. I watched as they danced, sang and listened to speeches by student activists and sympathetic professors. That first protest had an angry edge, but it was also completely peaceful and endlessly fascinating to a kid from a small town in Kentucky who’d never seen a drum circle before.

But things soon got worse, much worse. Protests got more unruly, and student activists got more aggressive. The entire campus was in a state of conflict. In Sedgwick’s words, students were “waging holy war on one another.” Small groups of students occupied administrative offices, and angry activists shouted down their political opponents in class and often attempted to intimidate them outside class. I was shouted down repeatedly, and twice I received disturbing handwritten notes in my campus mailbox in response to my anti-abortion advocacy. My student peers told me to “go die.”

Watching the protests and experiencing the shout-downs changed the course of my career. I was both enthralled by the power of protest and repulsed by the efforts to silence dissenters. Given the immense cultural influence of American higher education, I agreed with the Supreme Court’s famous words in the 1957 case Sweezy v. New Hampshire : “Teachers and students must always remain free to inquire, to study and to evaluate, to gain new maturity and understanding; otherwise, our civilization will stagnate and die.” Those words, combined with my own negative encounters at Harvard, helped define my legal career. From that point forward, I would defend free speech.

It’s been more than 30 years since that first campus protest, and over that time I’ve seen countless protests, I’ve defended countless protesters — and I’ve even been protested against at several schools. In the course of those cases and confrontations, I’ve learned that the issue of campus protest is remarkably complex and that campus culture is at least as important as law and policy in setting the boundaries of debate.

There is profound confusion on campus right now around the distinctions among free speech, civil disobedience and lawlessness. At the same time, some schools also seem confused about their fundamental academic mission. Does the university believe it should be neutral toward campus activism — protecting it as an exercise of the students’ constitutional rights and academic freedoms but not cooperating with student activists to advance shared goals — or does it incorporate activism as part of the educational process itself, including by coordinating with the protesters and encouraging their activism?

The simplest way of outlining the ideal university policy toward protest is to say that it should protect free speech, respect civil disobedience and uphold the rule of law. That means universities should protect the rights of students and faculty members on a viewpoint-neutral basis, and they should endeavor to make sure that every member of the campus community has the same access to campus facilities and resources.

That also means showing no favoritism among competing ideological groups in access to classrooms, in the imposition of campus penalties and in access to educational opportunities. All groups should have equal rights to engage in the full range of protected speech, including by engaging in rhetoric that’s hateful to express and painful to hear. Public chants like “Globalize the intifada” may be repugnant to many ears, but they’re clearly protected by the First Amendment at public universities and by policies protecting free speech and academic freedom at most private universities.

Still, reasonable time, place and manner restrictions are indispensable in this context. Time, place and manner restrictions are content-neutral legal rules that enable a diverse community to share the same space and enjoy equal rights.

Noise limits can protect the ability of students to study and sleep. Restricting the amount of time any one group can demonstrate on the limited open spaces on campus permits other groups to use the same space. If one group is permitted to occupy a quad indefinitely, for example, then that action by necessity excludes other organizations from the same ground. In that sense, indefinitely occupying a university quad isn’t simply a form of expression; it also functions as a form of exclusion. Put most simply, student groups should be able to take turns using public spaces, for an equal amount of time and during a roughly similar portion of the day.

Civil disobedience is distinct from First Amendment-protected speech. It involves both breaking an unjust law and accepting the consequences. There is a long and honorable history of civil disobedience in the United States, but true civil disobedience ultimately honors and respects the rule of law. In a 1965 appearance on “Meet the Press,” the Rev. Dr. Martin Luther King Jr. described the principle perfectly: “When one breaks the law that conscience tells him is unjust, he must do it openly, he must do it cheerfully, he must do it lovingly, he must do it civilly — not uncivilly — and he must do it with a willingness to accept the penalty.”

But what we’re seeing on a number of campuses isn’t free expression, nor is it civil disobedience. It’s outright lawlessness. No matter the frustration of campus activists or their desire to be heard, true civil disobedience shouldn’t violate the rights of others. Indefinitely occupying a quad violates the rights of other speakers to use the same space. Relentless, loud protest violates the rights of students to sleep or study in peace. And when protests become truly threatening or intimidating, they can violate the civil rights of other students, especially if those students are targeted on the basis of their race, sex, color or national origin.

The result of lawlessness is chaos and injustice. Other students can’t speak. Other students can’t learn. Teachers and administrators can’t do their jobs.

In my experience as a litigator , campus chaos is frequently the result of a specific campus culture. Administrators and faculty members will often abandon any pretense of institutional neutrality and either cooperate with their most intense activist students or impose double standards that grant favored constituencies extraordinary privileges. For many administrators, the very idea of neutrality is repugnant. It represents a form of complicity in injustice that they simply can’t and won’t stomach. So they nurture and support one side. They scorn the opposition, adopting a de facto posture that says , “To my friends, everything; for my enemies, the law.”

I’ve experienced this firsthand. I vividly remember representing a campus Christian group in a dispute at Tufts University in 2000 . The group had been derecognized for requiring that student leaders of their group share that group’s traditional sexual ethic, which reserves sex for heterosexual marriage. You might disagree strongly with that view, but granting religious groups the flexibility to impose faith-based requirements on religious leaders fits squarely within the American tradition of free exercise of religion.

Tufts is a private university, so it has some flexibility in suppressing religious expression on campus, but it had no excuse for attempting to toss a Christian group from campus at the same time that it permitted acts of intimidation against those Christian students. For example, at the most contentious moment of the dispute, Tufts officials prevented my student clients and me from entering the hearing room where their appeal was being heard, while a crowd of protesters gathered in a darkened hallway, pressed up around us and herded us into a corner of the hall. There was no campus outrage at this act of intimidation. We saw no administrative response.

University complicity in chaos isn’t unusual. In a case I worked on when I was president of the Foundation for Individual Rights and Expression, we discovered that administrators at Washington State University’s Pullman campus had actually helped plan a disruptive protest against a play put on by a student director, an intentionally provocative show that mocked virtually every group on campus.

University or faculty participation in unlawful protest isn’t confined to the cases I worked on. At Oberlin College, administrative facilitation of ugly and defamatory student protests outside a local business ultimately cost the school $36 million in damages. At Columbia, hundreds of sympathetic faculty members staged their own protest in support of the student encampment on the quad, and there are reports that other faculty members have attempted to block members of the media from access to the student encampment.

None of this is new. All of it creates a culture of impunity for the most radical students. Disruptive protesters are rarely disciplined, or they get mere slaps on the wrist. They’re hailed as heroes by many of their professors. Administrators look the other way as protesters pitch their tents on the quad — despite clear violations of university policy. Then, days later, the same administrators look at the tent city on campus, wring their hands, and ask, “How did this spiral out of control?”

There is a better way . When universities can actually recognize and enforce the distinctions among free speech, civil disobedience and lawlessness, they can protect both the right of students to protest and the rights of students to study and learn in peace.

In March a small band of pro-Palestinian students at Vanderbilt University in Nashville pushed past a security guard so aggressively that they injured him , walked into a university facility that was closed to protest and briefly occupied the building. The university had provided ample space for protest, and both pro-Israel and pro-Palestinian students had been speaking and protesting peacefully on campus since Oct. 7.

But these students weren’t engaged in free speech. Nor were they engaged in true civil disobedience. Civil disobedience does not include assault, and within hours the university shut them down. Three students were arrested in the assault on the security guard, and one was arrested on charges of vandalism. More than 20 students were subjected to university discipline, three were expelled , and one was suspended.

The message was clear: Every student can protest, but protest has to be peaceful and lawful. In taking this action, Vanderbilt was empowered by its posture of institutional neutrality . It does not take sides in matters of public dispute. Its fundamental role is to maintain a forum for speech, not to set the terms of the debate and certainly not to permit one side to break reasonable rules that protect education and safety on campus.

Vanderbilt is not alone in its commitment to neutrality. The University of Chicago has long adhered to the Kalven principles , a statement of university neutrality articulated in 1967 by a committee led by one of the most respected legal scholars of the last century, Harry Kalven Jr. At their heart, the Kalven principles articulate the view that “the instrument of dissent and criticism is the individual faculty member or the individual student. The university is the home and sponsor of critics; it is not itself the critic. It is, to go back once again to the classic phrase, a community of scholars.”

Contrast Vanderbilt’s precise response with the opposing extremes. In response to the chaos at Columbia, the school is finishing the semester with hybrid classes, pushing thousands of students online. The University of Southern California canceled its main stage commencement ceremony , claiming that the need for additional safety measures made the ceremony impractical. At both schools the inability to guarantee safety and order has diminished the educational experience of their students.

While U.S.C. and Columbia capitulate, other schools have taken an excessively draconian approach. Gov. Greg Abbott of Texas posted on X, “Students joining in hate-filled, antisemitic protests at any public college or university in Texas should be expelled.” On April 25 the Foundation for Individual Rights and Expression sent a forceful letter to the president of the University of Texas at Austin condemning the display of force on campus. “U.T. Austin,” it wrote, “at the direction of Gov. Greg Abbott, appears to have pre-emptively banned peaceful pro-Palestinian protesters due solely to their views rather than for any actionable misconduct.”

At Emory University, footage emerged of police tackling a female professor who posed no obvious danger to the police or anyone else. Protests are almost always tense, and there is often no easy way to physically remove protesters from campus, but the video footage of the confrontation with the professor was shocking. It’s hard to conceive of a justification for the violent police response.

At this moment, one has the impression that university presidents at several universities are simply hanging on, hoping against hope that they can manage the crisis well enough to survive the school year and close the dorms and praying that passions cool over the summer.

That is a vain hope. There is no indication that the war in Gaza — or certainly the region — will be over by the fall. It’s quite possible that Israel will be engaged in full-scale war on its northern border against Hezbollah. And the United States will be in the midst of a presidential election that could be every bit as contentious as the 2020 contest.

But the summer does give space for a reboot. It allows universities to declare unequivocally that they will protect free speech, respect peaceful civil disobedience and uphold the rule of law by protecting the campus community from violence and chaos. Universities should not protect students from hurtful ideas, but they must protect their ability to peacefully live and learn in a community of scholars. There is no other viable alternative.

The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips . And here’s our email: [email protected] .

Follow the New York Times Opinion section on Facebook , Instagram , TikTok , WhatsApp , X and Threads .

David French is an Opinion columnist, writing about law, culture, religion and armed conflict. He is a veteran of Operation Iraqi Freedom and a former constitutional litigator. His most recent book is “Divided We Fall: America’s Secession Threat and How to Restore Our Nation .” You can follow him on Threads ( @davidfrenchjag ).

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Violence against women on Twitter in India: Testing a taxonomy for online misogyny and measuring its prevalence during COVID-19

Roles Conceptualization, Data curation, Formal analysis, Methodology, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliation UNICEF, UNICEF HQ, New York, New York, United States of America

Roles Methodology, Supervision, Writing – review & editing

Affiliation Department of Computer Science, School of Engineering, University of California San Diego, San Diego, California, United States of America

Affiliation Center on Gender Equity and Health, Department of Medicine, University of California San Diego, San Diego, California, United States of America

Roles Supervision, Writing – review & editing

Affiliation Division of Health Promotion and Behavioral Science, School of Public Health, San Diego State University, San Diego, California, United States of America

Affiliation School of Social Work, San Diego State University, San Diego, California, United States of America

Roles Funding acquisition, Methodology, Supervision, Writing – review & editing

• Nabamallika Dehingia, 
• Julian McAuley, 
• Lotus McDougal, 
• Elizabeth Reed, 
• Jay G. Silverman, 
• Lianne Urada, 

• Published: October 25, 2023
• https://doi.org/10.1371/journal.pone.0292121
• Peer Review
• Reader Comments

Online misogyny is a violation of women’s digital rights. Empirical studies on this topic are however lacking, particularly in low- and middle- income countries. The current study aimed to estimate whether prevalence of online misogyny on Twitter in India changed since the pandemic.

Based on prior theoretical work, we defined online misogyny as consisting of six overlapping forms: sexist abuses, sexual objectification, threatening to physically or sexually harm women, asserting women’s inferiority, justifying violence against women, and dismissing feminist efforts. Qualitative analysis of a small subset of tweets posted from India (40,672 tweets) substantiated this definition and taxonomy for online misogyny. Supervised machine learning models were used to predict the status of misogyny across a corpus of 30 million tweets posted from India between 2018 and 2021. Next, interrupted time series analysis examined changes in online misogyny prevalence, before and during COVID-19.

Qualitative assessment showed that online misogyny in India existed most in the form of sexual objectification and sexist abusive content, which demeans women and shames them for their presumed sexual activity. Around 2% of overall tweets posted from India between 2018 and 2021 included some form of misogynistic content. The absolute volume as well as proportion of misogynistic tweets showed significant increasing trends after the onset of COVID-19, relative to trends prior to the pandemic.

Findings highlight increasing gender inequalities on Twitter since the pandemic. Aggressive and hateful tweets that target women attempt to reinforce traditional gender norms, especially those relating to idealized sexual behavior and framing of women as sexual beings. There is an urgent need for future research and development of interventions to make digital spaces gender equitable and welcoming to women.

Citation: Dehingia N, McAuley J, McDougal L, Reed E, Silverman JG, Urada L, et al. (2023) Violence against women on Twitter in India: Testing a taxonomy for online misogyny and measuring its prevalence during COVID-19. PLoS ONE 18(10): e0292121. https://doi.org/10.1371/journal.pone.0292121

Editor: Mahdi Zareei, Technologico de Monterrey, MEXICO

Received: November 8, 2022; Accepted: September 13, 2023; Published: October 25, 2023

Copyright: © 2023 Dehingia et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data necessary for replicating the study findings can be found here: https://doi.org/10.7910/DVN/H3IXDN .

Funding: This study was funded by the Bill and Melinda Gates Foundation Grant No. OPP1163682. The funding body had no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Violence against women (VAW) on online platforms is a violation of women’s digital rights. It can push women out of online spaces, and impact their social, economic, as well as health outcomes [ 1 ]. One of the most prevalent forms of online VAW, online misogyny or hate speech against women refers to any content that tries to threaten, intimidate, and shame women, or any rhetoric that emphasizes the authority of men over women [ 1 ]. According to a multi-country online survey in 2019, around two-thirds of women users of the internet reported having received some form of sexist or hateful language designed to attack or humiliate them [ 2 ]. Such experiences can often cause increased anxiety, depression, and lower self-esteem among women [ 3 , 4 ]. The extent and frequency of misogynistic attacks varies across different groups of women; younger women, and women belonging to marginalized racial or sexual identity groups are more at risk of experiencing online misogyny [ 2 , 5 – 9 ]. The past two years have thus noted an increasing number of calls, for hate speech regulation on social media platforms [ 10 , 11 ].

The majority of the existing research on online violence, including online misogyny, has focused on high-income countries, primarily due to greater digital access and use of social media platforms in these regions [ 12 ]. To our knowledge, no academic study has examined the characteristics of online misogyny in low- and middle-income countries (LMIC). This presents a key gap in literature, given that the use of social media, and consequently, the amount of online hateful content, has increased rapidly in LMICs in recent years [ 13 , 14 ]. India, in particular, is an important geography to examine online misogyny. Compared to many other LMICs, adverse gender norms and offline forms of gender-based violence are more prevalent in the country [ 15 , 16 ], putting women at a higher risk of experiencing violence in online spaces. Use of social media platforms is also on the rise in the country. As of January 2022, India had over 25 million active Twitter users, with a 30% increase in average daily users in 2020. Although Twitter has an urban bias; most Twitter users in the country are likely to be educated, living in cities, and of middle to high wealth status [ 17 ]. Nonetheless, Twitter is being frequently used as a key platform for feminist discussions and organizing in India [ 18 ], which can increase the likelihood of misogynist expression as a backlash [ 19 ]. The current study aims to contribute to the growing literature on online VAW on social media platforms, by measuring online misogyny on Twitter in India.

Limited global research suggests a potential increase in this form of violence since the pandemic, impacting millions of women daily [ 10 ]. An analysis of tweets and Facebook posts from South and South-east Asia by United Nations Women found 168 percent increase in misogynistic content during March-June 2020, when compared to the same period in 2019 [ 20 ]. However, despite many such analyses of the effect of COVID-19 on online content, little is known about the features and prevalence of online misogyny, and how it might have changed over the course of the pandemic. In this study, we take a first step to provide temporal estimates of online misogyny on Twitter in India from 2018 to 2021, and systematically examine any changes in its rates of increase/decrease before and during COVID-19.

As early as May 2020, the United Nations Secretary General made a global appeal to tackle COVID-19 influenced hate speech on digital platforms, with hateful content covering “ stereotyping , stigmatization , and the use of derogatory , misogynistic , racist , xenophobic , Islamophobic or antisemitic language ” [ 21 ]. Potential drivers of such hate speech during the pandemic include increased isolation due to stay-at-home orders, greater use of social media platforms, greater exposure to polarizing and differing views on the social media platforms, and fear, uncertainty, and anxiety of living through the pandemic [ 22 ]. Isolation is a strong motivational factor for hate speech on social media [ 23 ], and the confined living conditions, health and financial worries following the lockdown can create tension and stress, increasing the risk for online expression of hate speech. With regards to hate speech against women in particular, global evidence in support of increasing cases of gender-based violence during the pandemic further points to a potential increase in violence against women on online platforms [ 24 ]. An analysis of Twitter data before and after the onset of COVID-19 will allow us to test this hypothesis of increased levels of misogyny on Twitter since the pandemic.

Defining online misogyny

A key reason for the limited empirical evidence on online misogyny, is the lack of a standard taxonomy [ 25 ]. A majority of existing research on this topic has lacked theoretical considerations, classifying it as content that includes identifiable sexist slurs [ 26 – 28 ]. Earlier works on misogyny detection relied on lexicon-based methods, where misogynistic tweets were identified based on the presence of gendered abusive words [ 29 , 30 ]. Anzovino et. al (2018) introduced a benchmark dataset of tweets classified as misogynistic, with online misogyny defined as consisting of five forms: a) slurring, b) stereotypes, c) sexual harassment and threats, d) dominance to preserve men’s control, and e) derailing to justify VAW [ 31 , 32 ]. Multiple other studies have used this definition for misogyny detection [ 33 , 34 ]. We build on this taxonomy, by adapting it to the context of India, and locating it within a theoretical understanding of misogyny as a notion central to the feminist theory.

Feminist theory posits that misogyny, whether online or offline, stems from patriarchal values justifying men’s control over women via degradation and violence; misogyny actively seeks to silence women and maintain the status quo of patriarchal gender roles [ 35 ]. Sobieraj describes intimidating, shaming, and discrediting as the three key strategies often adopted by perpetrators of online misogyny, where attackers draw on women’s fear of sexual assault and physical violence to intimidate them, sexually objectify to publicly shame them, and discredit or dismiss their achievements as well as their fight to equality [ 1 ]. Asserting the inferiority of women, dismissing feminist movements, and objectifying women or viewing women primarily as an object of men’s desire, are thus at the heart of misogyny [ 36 ]. Our preliminary research on misogyny detection in South-Asian countries noted an emphasis on dismissal of feminist thought and discrediting of women’s claims of gender-based violence [ 37 ]. Guided by these theoretical and empirical works, we define online misogyny as any rhetoric or content that uses hostile and malicious language targeted at women, objectifies them, threatens them with physical or sexual harm, tries to assert their inferiority, justifies gender-based violence, or discredits feminist activists and their efforts. By qualitatively analyzing a large dataset of tweets (over 40,000 tweets), we first test whether this definition and taxonomy is valid for content posted on Twitter from India. This is followed by estimation of temporal online misogyny, using machine learning techniques that rely on the validated taxonomy. Our study objectives are two-fold: a) testing a taxonomy for online misogyny in India that is rooted in feminist theory, and b) using this taxonomy, estimate whether prevalence of online misogyny on Indian Twitter changed since the pandemic, across a sample of 30 million tweets posted between 2018 and 2021. Given the current lack of research on the extent and forms of misogyny on Twitter in LMICs, our study hopes to generate evidence that can be used to advocate for policies and strategies to address misogyny on Twitter.

Online misogyny taxonomy

We define online misogyny as consisting of six broad and overlapping forms: a) sexist abusive content, b) sexual objectification, c) threatening to physically or sexually harm women (presence of sexist language that indicates the threat being directed at women), d) asserting inferiority, e) justifying VAW, and f) dismissing feminist efforts. We provide definitions and example tweets for each form of online misogyny as Supporting Information. This taxonomy is guided by feminist theoretical work on misogyny [ 36 ], prior research on online misogyny detection, particularly the studies by Anzovino et. al (2018) , and our preliminary research on measuring online misogyny in South Asian countries.

Extracting geolocated tweets

We used geotagged English language tweets posted from India between January 2018 and December 2021 (30 million tweets). All tweets were extracted using the official Twitter API for academic research, which allows extraction of historical tweets. The most recent 800–1000 tweets published every hour from January 1 st 00:00 2018 to December 31 st 23:00 2021 were collected.

We used supervised machine learning models to classify tweets as misogynist/non-misogynist, as well as the six different forms of misogyny. Fig 1 depicts the different steps involved in our analysis. We describe each step in detail in the following sections.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0292121.g001

Qualitative analysis of tweets for misogyny

Supervised machine learning models require a ground truth dataset. To that end, two trained undergraduate students first qualitatively coded a small subset of tweets, as misogynist/non-misogynist, and the six forms of misogyny. The selection of the subset of tweets was done using two approaches. First, from the large corpus of geotagged tweets between 2018 and 2020, we extracted posts containing representative keywords frequently used to harass and abuse women. To identify keywords, we contacted authors of an academic study on hate speech detection on Twitter. They used a list of keywords for different types of hate speech on Twitter, including misogyny [ 38 ]. We added a few words relevant to the Indian context to this list. Identified words included abusive terms, as well as generic words related to gender such as " woman ", " feminism " etc. It is key to note here that there may be certain tweets that contain abusive words, but might be non-offensive in their use and overall meaning. The inclusion of all such tweets that contained abusive words allowed us to build a ground truth dataset that identified the nuances in the use of abusive words. With this process, we selected a subset of 35,672 tweets.

Next, we chose a random subset of 5,000 tweets from the large corpus of geotagged tweets from 2018–2020. This dataset was included to ensure that tweets unrelated to women or gender were also represented in our ground truth sample. The main goal of using the two separate approaches to build the ground truth dataset was to collect a set of tweets that would have an adequate representation of both misogynist and non-misogynist tweets. A total of 40,672 tweets were thus selected for qualitative coding, with these two processes.

The qualitative coding of the tweets was carried out in two phases. First, the subset of tweets was coded as misogynist or non-misogynist by five trained undergraduate students. Every tweet was classified by at least two coders. Inter-rater kappa scores were calculated to examine reliability of the coding [ 39 ]. Any disagreement between the coders was resolved by the lead author of the study. Next, all tweets classified as misogynist were coded by two graduate students experienced in gender research, as the six forms of online misogyny. Inter-rater kappa scores were calculated for this step as well. The qualitative coding of misogynist tweets provided us with the opportunity to validate our taxonomy of misogyny. The two researchers did not identify any content that was outside the purview of the six predefined categories.

Selection of features or independent variables

Tweets were cleaned before running any machine learning model. This included deleting urls and ignoring punctuation and case. We also removed stop words, which are commonly used words such as “is”, “the”, “a” etc., from our text data. The decision to drop stop words was made after running multiple models with and without stop words. Next, the data was processed with text vectorization , in order for the machine learning regression models to mathematically interpret it. Text vectorization refers to the process of creating features or input variables in a model (independent variables in public health terminology). There are multiple methods for generating features or independent variables from text data. We used three types of feature selection methods: a) term frequency-inverse document frequency (TF-IDF) unigrams, b) TF-IDF unigrams and bigrams, and c) TF-IDF unigrams, bigrams, and linguistic features (number of words and number of adjectives). These methods were chosen based on guidelines suggested by a previous study on abuse detection [ 32 ]. TF-IDF is weighted word frequency, and reflects how important a word is to a document in a collection or corpus [ 40 ]. While TF-IDF with unigrams concerns with single words only, TF-IDF with unigrams and bigrams includes sequences of two words along with singular words. In addition to weighted word frequencies, linguistic features such as length of the tweet or number of adjectives in a tweet can also provide meaningful information about the content of the tweet. It is possible that tweets that are misogynist are shorter in length and have a relatively higher number of adjectives (abusive words or sentiments).

Running supervised machine learning models and predicting misogyny: Train and test models for best fit

We randomly split our data into a train and test dataset in a 70:30 ratio. The training dataset is used to train the machine learning model, while the test dataset is used to assess the performance of the trained model. To estimate the hyperparameters to be specified in the training model, we used a k-fold validation process. In this method, the training data are partitioned into k subsets of approximately equal size and one of the subsets becomes the validation set where the hyperparameters are validated. The remaining k-1 subsets are used as training data and this process is repeated.

We ran five different types of supervised machine learning models to classify tweets as misogynist/non-misogynist: a) Naïve Bayes (NB), b) Support Vector Machine (SVM), c) Ridge logistic regression, d) Multi-layer Perceptron Neural Network (MPNN), and e) Bidirectional Encoder Representations from Transformers (BERT). Naïve Bayes is a probabilistic model based on Bayes theorem and considers a strong independence assumption [ 41 ]. SVMs are based on a structural risk minimization principle, which aims to test a hypothesis for which we can guarantee the lowest true error. SVMs aim to create a hyperplane ( maximum-marginal hyperplane ), which separates the categories of the outcome label [ 42 ]. MPNN are complex non-linear models with a hierarchical or multi-layered structure. These models have been commonly used in text classification tasks [ 32 ]. BERT is a pre-trained language model (BERT is pretrained on a large corpus of English texts from Wikipedia and BookCorpus), with transformer architecture [ 43 ]. Transformers learn contextual relations between words; they are deep learning models where each output element is connected to every input element and the weightings between them are dynamically calculated. While previous models examine text sequences in a single direction (left-to-right or right-to-left training), BERT’s key difference and innovation is the application of bidirectional training. The BERT Transformer encoder reads the entire sequence of words at once. Specifically, we used the bert-large-uncased version, which consists of 24 layers (1024 hidden dimensions), 16 attention heads, and a total of 340M parameters. The Transformers library in Python was used to implement this approach.

The performance of each model was evaluated on the test set, by comparing actual labels or outcome variable values with the values predicted by the trained model. We used two evaluation metrics: (a) F1 score and (b) receiver operating characteristic—area under the curve (AUC). The receiver operating characteristic captures precision and recall at all potential decision thresholds, so the area under it is an appropriate metric to measure overall model performance. F1 score is the harmonic mean of precision and recall. The models with the best F1 and AUC estimates were chosen to predict the labels/outcome variable values for each tweet in our original dataset of over 30 million geotagged tweets posted between January 2018 to December 2021.

The analysis was repeated with each of the six forms of online misogyny as the outcome variable or label, with the sub-sample of tweets that was classified misogynist (N = 3894).

Analysis of misogyny prevalence pre and during COVID-19

Y t is the proportion of online misogyny related tweets, T is the time since COVID-19 related lockdown has been implemented in India (in days), X t is the dummy variable representing the imposition of lockdown, and X t T is an interaction term. β 1 represents the underlying pre-COVID-19 trend in misogyny prevalence, or the change in misogynistic tweets for every one unit time increase. β 2 captures the immediate effect of the onset of COVID-19, and β 3 represents the slope change in the proportion of misogynistic tweets following COVID-19, in comparison to trends prior to COVID-19. We ran interrupted time series regression models with two separate outcomes: a) total number of misogynistic tweets and b) proportion of misogynistic tweets per 100000 tweets.

Qualitative assessment of misogyny across tweets

Around 8% of the tweets that were qualitatively analyzed were classified as misogynist (N = 40,672). Among those identified as misogynist, 57% were related to sexual objectification, 34% covered sexist abusive content, 6% justified violence against women, and 5% were content that dismissed feminism or feminist efforts. A small proportion of misogynist tweets included threats of harm (1%) and assertion of authority (1%). The two researchers who conducted the qualitative coding did not identify any content that was outside the purview of the six predefined categories of online misogyny. The coding had good reliability, with an inter-rater kappa score of 0.87.

The tweets classified as sexual objectification (57%) included text that sexualized women’s body, shamed women for presumed sexual activity ( slut shaming ), and text that included lewd remarks and sexual solicitation. These tweets emphasized the treatment of women as sexual objects of men’s sexual desire. The category “justify violence against women” (6%) included tweets that highlighted the prevalence of rape myth acceptance and victim blaming among Twitter users. Tweets aggressively asserted that women needed to act a certain way to reduce the risk of rape and violence (e.g., not wear short clothes, not go out at night). This category also included tweets that aimed to derail the conversation from VAW, by focusing on “fake rape cases” in India. Around 10% of these tweets included the hashtag “#mentoo”, with tweets noting men’s victimization with fake rape allegations. Additionally, there were a small number of tweets (n = 3) that dismissed marital rape. For example, a tweet stated that all married women were free to be raped by their husbands.

The category on dismissal of feminist efforts included a number of tweets declaring feminism as a “virus”, with 13% of the tweets including the word “feminazis”. Around 16% of the tweets classified as dismissal of feminist efforts were also categorized as justifying VAW.

Machine learning models to predict misogyny

We ran the different types of machine learning models, separately for each outcome variable: misogyny, sexual objectification, sexist abusive content, justifying VAW, dismissing feminist efforts, threatening to harm, and asserting authority. For the analysis with misogyny as the outcome variable, BERT performed the best [ Table 1 ]. This was followed by the logistic regression models (with TF-IDF unigrams). The Naïve Bayes had the lowest values of AUC as well as F1 score, for misogyny prediction.

https://doi.org/10.1371/journal.pone.0292121.t001

Predicting online misogyny across tweets from 2018–2021 and examining prevalence

We used the specifications from our best performing model to predict misogyny across the large Twitter dataset of 30 million tweets. Each tweet in this dataset was classified as misogyny or non-misogyny by this model. Overall, 1.6% of the geotagged tweets collected from India included misogynist content.

We ran separate interrupted time series models with total number of misogynistic tweets, and proportion of misogynistic tweets per 100000 tweets as the outcomes ( Table 2 ). The volume of misogynistic tweets increased significantly by 30 times immediately after the onset of COVID-19, with increasing trends in the following months between April 2020-December 2021. The proportion of misogynistic tweets per 100000 tweets also showed significant and sustained increasing trends after the onset of COVID-19, relative to trends prior to the pandemic. However, there was a significant decrease immediately after the onset of COVID-19, followed by the increasing trends.

https://doi.org/10.1371/journal.pone.0292121.t002

Digital social media portals are often signaled to be democratic public spaces. However, we find that similar to the offline world, they are home to increasing gender inequalities. Our study observes online misogyny on Twitter in India to be prevalent, with increasing trends since COVID-19. Around 2% of overall daily tweets between 2018 and 2021 included some form of misogynistic content. This translates to millions of misogynistic tweets every day, given that around 20 billion tweets are posted daily on average. Our study draws attention to online misogyny as a topic for gender research in India, a country where internet use is on the rise and offline forms of gender-based violence are highly prevalent.

We find evidence in support of increasing trends in overall volume as well as proportion of misogynistic content on Twitter since the pandemic. However, our results show an immediate drop in the proportion of misogynistic tweets after the pandemic, followed by sustained and significant growth in the next two years. This could be because of an increase in tweets related to other relevant topics such as COVID-19, health services, and vaccines, right after the onset of the pandemic. Nonetheless, the sustained increasing trend after the initial drop is worrisome, and warrants further assessments of long-term changes. Our study contributes to the existing literature that shows significant increases in other offline forms of VAW since the pandemic.

We find that online misogyny exists in the form of sexual objectification, sexist abusive content, content that threatens to harm, asserts authority, justifies VAW, and dismisses feminism, with sexual objectification and sexual abuse of women being the most common forms. Multiple prior studies have shown online sexual harassment and victimization to be associated with adverse mental and psychological consequences, particularly for young girls [ 44 , 45 ]. Online sexual harassment is unique compared to offline experiences in that individuals who are not directly attacked by the abusers/social media users, are also exposed to the sexually charged posts. In addition to causing mental distress, for young users of the Internet, exposure to such content might cause endorsements of patriarchal beliefs that view women as sexual beings and normalize VAW, leading to a continuation of the cycle of violence [ 46 , 47 ]. While findings are related to the online world, this is indicative of an ongoing acceptability of sexual harassment in the country [ 48 ]. We find that sexual objectification related tweets also include content that shame women for their real or presumed sexual activity. Our findings show that online abusers often use women’s sexual autonomy as a weapon for demeaning and disrespecting them, further perpetuating adverse gender norms that deny women sexual freedom.

Another manifestation of online misogyny is the dismissal of feminist efforts, and justification of VAW. Resistance to advancement of gender equality is a common feature of the feminist struggle, and like the offline world, it is evidently present in the Indian Twitter space. This ‘backlash’ effect has been documented for offline movements, and it can take different forms such as denial of the issue, derailing of conversation, rejection of men’s responsibility, and discrediting of feminist activities [ 49 ]. Prior research has highlighted the disproportionate amount of online misogyny and violence directed at women politicians, journalists, and women engaging in feminist debate on Twitter. Our analysis did not classify tweets that were directed at specific women, but we found many posts that included words such as “ feminazis ”, and dismissed feminism as a “ problem ” and a “ virus ”. It is likely that such posts were directed at individuals who engage in conversations related to feminism or gender equality. Denial of the problem by focusing on fake rape allegations was another common theme across tweets categorized as justification of VAW. Such content, in addition to spreading disinformation, has the potential to widen digital gender inequalities by acting as triggering content for survivors of gender-based violence.

Currently, there are no dedicated legislations in India against gender-based cyber violence [ 50 ]. The laws protecting citizens from cyber hate do not specifically recognize sexist or misogynistic trolling. Our findings show that along with strengthening of laws and regulations to prevent online misogyny, shifting entrenched gender norms will be critical to achieving gender equality in digital spaces. Norms-based interventions addressing gender-based violence in general can consider inclusion of elements of digital violence in their programs.

In addition to providing insights on the prevalence and characteristics of online misogyny, our work is one of the first attempts to detect misogyny across a very large dataset of tweets from India. We tested multiple machine learning models, and found the pre-trained language model, BERT, to be the best performing models. However, our analysis is limited to English tweets only, and does not cover tweets in any Indian vernacular language. Future studies should consider building multilingual models to provide improved insights related to misogyny on Twitter in India.

Our study has a few additional limitations. First, our findings are likely to have an urban bias, representative of the English-speaking population, since our data was limited to English tweets only. Next, this study only includes geotagged tweets from India, which constitutes around 2% of the overall tweets. However, few prior studies have noted that the large number of tweets created daily ensure the adequateness of geotagged tweets in being a good representation of overall Twitter conversations [ 51 , 52 ]. Third, we examine the changes in prevalence of misogyny before and after the start of the pandemic. We do not take into account other relevant key events that might trigger changes in the amount of misogynistic expression. Finally, our study focusses solely on the misogyny directed against women in India, and does not cover transmisogyny or different intersectional categories as relates to race, ethnicity, age, and sexual orientation. Future research should consider including these aspects, given that younger individuals, and individuals belonging to minority groups are more likely to experience violence online [ 5 ].

Conclusions

The past couple of years have noted increasing activism against the growing levels of misogyny on Twitter as well as other social media platforms. Our findings emphasize the need for inclusion of digital violence in the broader policy discourse on gender-based violence; online misogyny is a continuum of violence experienced by women in their offline worlds in India. Aggressive and hateful tweets that target women in the country attempt to reinforce traditional gender norms, especially those relating to idealized sexual behavior and framing of women as sexual beings. Our study provides evidence on increasing trends of misogyny on Indian Twitter, and emphasizes the utility of machine learning methods in examination of aspects related to VAW, which can support future research and development of interventions to make these digital spaces equitable and welcome for women.

Supporting information

S1 file. annotation guidelines/manual for undergraduate student coders..

https://doi.org/10.1371/journal.pone.0292121.s001

S2 File. Definition and example tweet for each category of online misogyny.

https://doi.org/10.1371/journal.pone.0292121.s002

Acknowledgments

We are grateful to Dr. James Fowler who supported us in accessing the Twitter data. We thank Riley Saham, Mairen Oates, Aviram Raj-Silverman, Lucas Fowler, and Max Goldberg, for serving as qualitative coders of the Twitter dataset. We are grateful to Wendy Wei Cheung for carrying out the qualitative coding of misogynist tweets.

• 1. Sobieraj S. Credible threat: Attacks against women online and the future of democracy: Oxford University Press; 2020.
• 2. The Economist Intelligence Unit (EIU). Measuring the prevalence of online violence against women New York: The Economist Intelligence Unit; 2020 [ https://onlineviolencewomen.eiu.com/ .
• 3. Amnesty International. Amnesty reveals alarming impact of online abuse against women: Amnesty International; 2017 [ https://www.amnesty.org/en/latest/news/2017/11/amnesty-reveals-alarming-impact-of-online-abuse-against-women/ .
• 4. Amnesty TT. A Toxic Place for Women, 2018.
• 5. deSHAME P. Young people’s experiences of online sexual harassment. ChildNet; 2017.
• View Article
• Google Scholar
• 7. Amnesty I. Toxic Twitter: a Toxic Place for Women. Report. 2018.
• 9. Pew Research Center. Pew Report on Online Harassment 2017. 2017.
• 10. UN Women. Online and ICT facilitated violence against women and girls during COVID-19. 2020.
• 11. Wang S, Affoum N. Cyber harassment: A growing concern in the age of COVID. In: World Bank, editor. 2021.
• 13. Jeremy Liebowitz, Geoffrey Macdonald, Vivek Shivaram, Vignaraja S. The Digitalization of Hate Speech in South and Southeast Asia: Conflict-Mitigation Approaches. Georgetown University of International Affairs2021.
• 14. Minority Rights. South Asia State of Minorities Report 2021 –Hate Speech against Minorities. South Asia Collective; 2022.
• 15. National Crime Records B. Crime in India 2016: Statistics. Ministry of Home Affairs, Government of India New Delhi; 2016.
• 16. IIPS & ICF. National Family Health Survey (NFHS-4) 2015–16. Mumbai, India; 2017.
• 17. Ananya Bhattacharya. In India, the internet is a place for cityfolk and men. Quartz India. 2018.
• 20. UN Women. ELIMINATING ONLINE HATE SPEECH TO SECURE WOMEN’S POLITICAL PARTICIPATION. 2021.
• 21. United Nations. United Nations Guidance Note on Addressing and Countering COVID-19 related Hate Speech. United Nations; 2020.
• PubMed/NCBI
• 23. Von Behr I. Radicalisation in the digital era: The use of the internet in 15 cases of terrorism and extremism. 2013.
• 27. National Democratic Institute. Tweets That Chill: Analyzing Online Violence Against Women in Politics. 2019.
• 30. Hewitt S, Tiropanis T, Bokhove C, editors. The problem of identifying misogynist language on Twitter (and other online social spaces)2016.
• 32. Anzovino M, Fersini E, Rosso P, editors. Automatic identification and classification of misogynistic language on twitter2018: Springer.
• 35. Becker M. Patriarchy and inequality: Towards a substantive feminism. U Chi Legal F. 1999:21.
• 36. Manne K. Down girl: The logic of misogyny: Oxford University Press; 2017.
• 37. Dehingia N, Lundgren R, Dey AK, Raj A. Trends in online misogyny before and during the COVID-19 pandemic: Analysis of Twitter data from five South-Asian countries. Center on Gender Equity and Health2021. p. 1–4.
• 38. Siegel A, Sterling J, Pullen B, Bonneau R, Nagler J, Tucker J. Trumping Hate on Twitter? Online Hate Speech and White Nationalist Rhetoric in the 2016 US Election Campaign and its Aftermath. 2017.
• 41. Dey L, Chakraborty S, Biswas A, Bose B, Tiwari S. Sentiment analysis of review datasets using naive bayes and k-nn classifier. arXiv preprint arXiv:161009982. 2016.
• 42. Joachims T, editor Text categorization with support vector machines: Learning with many relevant features. European conference on machine learning; 1998: Springer.
• 43. Devlin J, Chang M-W, Lee K, Toutanova K. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:181004805. 2018.
• 50. Anita Gurumurthy, Chami N. Digital India as if women matter: A policy discussion paper. India Office of the Heinrich Böll Foundation; 2018.

People who disrupt UW-Madison commencement could face suspension or even arrest

The University of Wisconsin-Madison is warning of disciplinary action and even arrest for people who disrupt its commencement ceremony this weekend.

The notice comes as campuses nationwide are roiled by protests over the Israel-Hamas war. Ceremonies at two other Big Ten schools last weekend were marked by pro-Palestinian protests. A few schools, including Columbia University, have canceled the event altogether.

UW-Madison doctoral and medical students graduate Friday at the Kohl Center. Undergraduates, master's and law students have their event at Camp Randall on Saturday. Altogether, UW-Madison anticipates nearly 8,600 students will be earning degrees.

A new disclaimer on the UW-Madison commencement website reads: "Disruption of university events, including graduation activities, violates state law and will not be tolerated."

"Considering recent events on our campus and on campuses around the country, we are attempting to set clear expectations for graduates and guests at commencement," UW-Madison spokesperson Kelly Tyrrell said.

Here's what to know:

Suspension, arrest on the table for commencement protesters

The notice says that, at a minimum, UW-Madison will recommend suspension for students and current graduates who disrupt commencement. The university will also place holds on degrees, registration and official university records.

Nonacademic suspensions last up to two years but only appear on a student's transcript during the suspension. Suspended students cannot be on campus during the suspension but can later re-enroll.

Students and non-students who engage in disruption "also risk arrest, citation, and criminal charges," the website said.

New security measures in place

UW-Madison declined to share details on its security plans but acknowledged its plans have been modified to increase public safety and reduce the likelihood of disruption. Nothing related to the ceremony program has changed, Tyrrell said. New safety measures include:

• Metal detectors, which are new for the Camp Randall ceremony but not the Kohl Center.
• Attendees are allowed one small clutch or a clear tote bag.
• All items and attendees are subject to search. Graduates will be asked to open their gowns for inspection as they enter.

Signs, noisemakers, flags and banners are banned, but the restrictions are nothing new. The items were prohibited in previous years, too.

"These modifications will change how people usually experience commencement but will be very familiar to anyone who has attended a home Badger football game," Tyrrell said.

Protesters mum on plans

It's unclear whether protesters who pitched tents on the Library Mall are making plans to disrupt the UW-Madison ceremonies. Organizers did not respond to questions sent by text message and Instagram.

At UW-Milwaukee, commencement is scheduled for May 19 at the UWM Panther Arena. Protester Kayla Patterson said a decision on whether to disrupt the event depends on the status of negotiation meetings.

"I can't really say specifically if we will or not, but it's not off the table," she said.

Class of 2024 includes many COVID-19 era high school grads

The class of 2024 graduated from high school in spring 2020. Most high school students that spring saw their ceremonies canceled or significantly curtailed because of the pandemic. Their freshman year of college was also severely disrupted.

The missed milestones has many students hoping for the traditional ceremony this time around.

A group of Jewish students requested UW-Madison ensure commencement go on as planned and discipline disrupters, according to a document shared with the Journal Sentinel. It was one of several demands made by the group, which reported feeling intimidated and harassed by pro-Palestinian protesters on or near the encampment.

Protests disrupted a few other university commencements

At the University of Michigan in Ann Arbor, some graduates interrupted the ceremony with chants, Palestinian flags and banners .

At Indiana University, a group staged a walkout during the president's speech and a "LET GAZA LIVE!" banner flew overhead.

Contact Kelly Meyerhofer at  [email protected] or 414-223-5168. Follow her on X (Twitter) at  @KellyMeyerhofer .